42
Temporal Evolution of the Lithospheric Mantle beneath the Eastern North China Craton ZHU-YIN CHU 1 *, FU-YUAN WU 1 , RICHARD J. WALKER 2 , ROBERTA L. RUDNICK 2 , LYNNETTE PITCHER 2 , IGOR S. PUCHTEL 2 , YUE-HENG YANG 1 AND SIMON A. WILDE 3 1 STATE KEY LABORATORY OF LITHOSPHERIC EVOLUTION, INSTITUTE OF GEOLOGY AND GEOPHYSICS, CHINESE ACADEMY OF SCIENCES, BEIJING 100029, CHINA 2 ISOTOPE GEOCHEMISTRY LABORATORY, DEPARTMENT OF GEOLOGY, UNIVERSITY OF MARYLAND, COLLEGE PARK, MD 20742, USA 3 DEPARTMENT OF APPLIED GEOLOGY, CURTIN UNIVERSITY OF TECHNOLOGY, PO BOX U1987, PERTH, W.A. 6845, AUSTRALIA RECEIVED JANUARY 28, 2009; ACCEPTED JULY 28, 2009 ADVANCE ACCESS PUBLICATION SEPTEMBER 3, 2009 It is now well established that lithospheric thinning of the eastern North China Craton (NCC) occurred during the Phanerozoic.The cause and extent of the thinning, however, remains highly debated. In this study, mantle xenoliths from the Paleozoic Mengyin kimber- lites, along with xenoliths from the Cenozoic Penglai and Shanwang basalts in Shandong Province, are investigated via tradi- tional petrographic and elemental, Sr^Nd^Hf^Os isotopic and pla- tinum-group element (PGE) analyses. Late Archean Os model ages of c. 2 5 Ga for the Mengyin peridotites provide confirmation that refractory, Archean lithospheric mantle existed beneath the east- ernmost portion of the NCC during the Paleozoic. Some Paleoproterozoic lithospheric mantle fragments may also be present in the Mengyin xenolith suite. In contrast, the spatially associated Penglai and Shanwang peridotite xenoliths are more fertile, and have Sr^Nd isotopic compositions similar to the depleted mantle. They differ dramatically from the Archean peridotites sampled by the Mengyin kimberlites. Osmium model ages for single samples range from mid-Proterozoic to modern, similar to variations observed in Phanerozoic convecting upper mantle as sampled by modern abys- sal peridotites. This suggests that the present lithospheric mantle beneath the eastern NCC formed in the Phanerozoic, despite the fact that Os model ages extend back to the Proterozoic. Some samples from the Penglai suite yield Proterozoic Lu^Hf clinopyroxene min- eral isochron ages that are consistent with the Os model ages, suggesting that Hf isotopes in modern convective upper mantle can preserve evidence for ancient melt depletion, similar to Os isotopes. Our results are consistent with thinning of the eastern NCC as a result of foundering of the deep crust and lithospheric mantle, but are inconsistent with stretching or refertilization models, as remnants of Archean mantle are expected to be present in these scenarios.The match between 187 Os/ 188 Os in convective upper mantle and the east- ern China lithopspheric mantle also precludes models that seek to explain the present lithosphere as being due to lateral translation of Proterozoic lithosphere. KEY WORDS: Cenozoic; isotopes; lithospheric mantle; Sm^Nd, Rb^Sr, Lu^Hf; North China Craton; Paleozoic; platinum-group elements (PGE); Re^Os INTRODUCTION Many studies have shown that a significant part of the original lithospheric mantle beneath the eastern portion of the North China Craton (NCC) was removed during the Phanerozoic (Fan & Menzies, 1992; Menzies et al ., 1993, 2007; Griffin et al ., 1998; Menzies & Xu, 1998; Xu, 2001). Early Paleozoic ( 470 Ma) diamondiferous *Corresponding author. Telephone: þ86-10-82998586. Fax: þ86-10-62010846. E-mail: [email protected] ß The Author 2009. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@ oxfordjournals.org JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 PAGES 1857^1898 2009 doi:10.1093/petrology/egp055

Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Temporal Evolution of the Lithospheric Mantlebeneath the Eastern North China Craton

ZHU-YIN CHU1 FU-YUAN WU1 RICHARDJWALKER2ROBERTA L RUDNICK2 LYNNETTE PITCHER2IGOR S PUCHTEL2 YUE-HENG YANG1 AND SIMON AWILDE3

1STATE KEY LABORATORY OF LITHOSPHERIC EVOLUTION INSTITUTE OF GEOLOGY AND GEOPHYSICS CHINESE

ACADEMY OF SCIENCES BEIJING 100029 CHINA2ISOTOPE GEOCHEMISTRY LABORATORY DEPARTMENT OF GEOLOGY UNIVERSITY OF MARYLAND COLLEGE PARK

MD 20742 USA3DEPARTMENT OF APPLIED GEOLOGY CURTIN UNIVERSITY OF TECHNOLOGY PO BOX U1987 PERTH WA 6845

AUSTRALIA

RECEIVED JANUARY 28 2009 ACCEPTEDJULY 28 2009ADVANCE ACCESS PUBLICATION SEPTEMBER 3 2009

It is now well established that lithospheric thinning of the eastern

North China Craton (NCC) occurred during the PhanerozoicThe

cause and extent of the thinning however remains highly debated

In this study mantle xenoliths from the Paleozoic Mengyin kimber-

lites along with xenoliths from the Cenozoic Penglai and

Shanwang basalts in Shandong Province are investigated via tradi-

tional petrographic and elemental Sr^Nd^Hf^Os isotopic and pla-

tinum-group element (PGE) analyses Late Archean Os model ages

of c 25 Ga for the Mengyin peridotites provide confirmation

that refractory Archean lithospheric mantle existed beneath the east-

ernmost portion of the NCC during the Paleozoic Some

Paleoproterozoic lithospheric mantle fragments may also be present

in the Mengyin xenolith suite In contrast the spatially associated

Penglai and Shanwang peridotite xenoliths are more fertile and

have Sr^Nd isotopic compositions similar to the depleted mantle

They differ dramatically from the Archean peridotites sampled by

the Mengyin kimberlites Osmium model ages for single samples

range from mid-Proterozoic to modern similar to variations observed

in Phanerozoic convecting upper mantle as sampled by modern abys-

sal peridotites This suggests that the present lithospheric mantle

beneath the eastern NCC formed in the Phanerozoic despite the fact

that Os model ages extend back to the Proterozoic Some samples

from the Penglai suite yield Proterozoic Lu^Hf clinopyroxene min-

eral isochron ages that are consistent with the Os model ages

suggesting that Hf isotopes in modern convective upper mantle can

preserve evidence for ancient melt depletion similar to Os isotopes

Our results are consistent with thinning of the eastern NCC as a

result of foundering of the deep crust and lithospheric mantle but

are inconsistent with stretching or refertilization models as remnants

of Archean mantle are expected to be present in these scenariosThe

match between 187Os188Os in convective upper mantle and the east-

ern China lithopspheric mantle also precludes models that seek to

explain the present lithosphere as being due to lateral translation of

Proterozoic lithosphere

KEY WORDS Cenozoic isotopes lithospheric mantle Sm^Nd Rb^Sr

Lu^Hf North China Craton Paleozoic platinum-group elements

(PGE) Re^Os

I NTRODUCTIONMany studies have shown that a significant part of theoriginal lithospheric mantle beneath the eastern portionof the North China Craton (NCC) was removed duringthe Phanerozoic (Fan amp Menzies 1992 Menzies et al1993 2007 Griffin et al 1998 Menzies amp Xu 1998 Xu2001) Early Paleozoic (470 Ma) diamondiferous

Corresponding author Telephone thorn86-10-82998586Faxthorn86-10-62010846 E-mail zhychumailigcasaccn

The Author 2009 Published by Oxford University Press Allrights reserved For Permissions please e-mail journalspermissionsoxfordjournalsorg

JOURNALOFPETROLOGY VOLUME 50 NUMBER10 PAGES1857^1898 2009 doi101093petrologyegp055

kimberlites in Mengyin Shandong Province and FuxianLiaoning Province indicate the presence of lithospherewith a thickness of 200 km at the time of volcanism(Menzies et al 1993 Griffin et al 1998 Wang et al 1998Menzies amp Xu 1998) (Fig 1) Nevertheless geophysicaldata and petrological studies of mantle xenoliths collectedfrom Cenozoic basalts suggest that the present lithosphereis much thinner (60^120 km) and considerably more fertile(Menzies et al 1993 Griffin et al 1998 Menzies amp Xu1998 Fan et al 2000 Xu 2001 Rudnick et al 2004 Chenet al 2006) Because of these extensive studies the easternNCC is now recognized as the best example of a reacti-vated craton with decoupled crust and mantle lithosphere(Carlson et al 2005 Menzies et al 2007)Despite extensive investigation the cause of lithospheric

thinning beneath the eastern block of the NCC remainshighly debated One school of thought suggests that theoriginal Archean lithospheric mantle and associated lowercrust were removed via density foundering and werereplaced by peridotite residues formed by melting of con-vecting upper mantle during the Mesozoic to earlyCenozoic (Gao et al 2002 2004 2008 Wu et al 20032005a) Another model suggests that mantle with

Proterozoic Os model ages originated by mixing the origi-nal Archean mantle lithosphere with juvenile astheno-spheric mantle via melt^peridotite interaction (Zhang2005 Zhang et al 2008) Others have proposed that thismantle lithosphere might be exotic derived from southernChina or elsewhere and has no relationship with thecrustal^mantle evolution of the NCC (Wu et al 2006)Mapping of spatial variations in mantle melt depletion

ages of the lithospheric mantle underlying the NCCbefore and after the loss of the deep lithosphere is essentialto diagnose the cause of the loss (Carlson et al 2005Foley 2008) However despite much recent study the agestructure of the lithospheric mantle underlying the NCCremains unclear partly because of the paucity of data forappropriate xenolith locations This is especially true forthe easternmost portion of the NCC To date Os isotopicdata have been reported for peridotite xenoliths from thePaleozoic Mengyin Fuxian and Tieling kimberlites Thesedata reveal the existence of Archean lithospheric mantlebeneath the eastern NCC during the early Paleozoic (Gaoet al 2002Wu et al 2006 Zhang et al 2008) In contrastperidotite xenoliths from Cenozoic basalts in QixiaLonggang and Kuandian (Fig 1) show considerable

Fig 1 Map of the North China Craton showing the xenolith localities discussed in the text Tectonic subdivisions are based on Zhao et al(2005) Inset shows location of the North China Craton (NCC) relative to other cratonic blocks (eg South China Craton) and interveningfold belts

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1858

variations in Os model ages These model ages range fromProterozoic to Phanerozoic and have been interpreted indifferent ways Most importantly no peridotite xenolithswith Archean Os model ages have been documented inthe Cenozoic basalts from the eastern NCC (Gao et al2002Wu et al 2003 2006)With the intent of providing a much more detailed pic-

ture of the age structure of the mantle lithosphere underly-ing the eastern portion of the NCC we have determinedthe petrography major and trace element chemistry Sr^Nd^Hf^Os isotope and platinum-group element (PGE)concentrations for two large suites of peridotite xenolithsfrom the Cenozoic Penglai and Shanwang basalts inShandong Province These data complement publisheddata for the nearby Qixia peridotites (Gao et al 2002Rudnick et al 2004)To bolster limited earlier data and fur-ther constrain the nature of the lithospheric mantle presentduring the Paleozoic we also analyzed additional mantlexenoliths from the Paleozoic Mengyin kimberlites

GEOLOGICAL BACKGROUNDAND SAMPLE DESCR IPT IONSEastern China is composed of the three main lithotectonicdomains the Central Asian Orogenic Belt and the NCCin the north the Dabie^Sulu ultrahigh-pressure collisionalbelt in the central region and the South China Craton(including the Yangtze Craton and Southeastern Chinaorogenic belt) in the south (Fig 1) The NCC is the oldesttectonic unit in China containing crust as old as 3800Ma (Liu et al 1992 Wu et al 2008) Neodymium TDM

model ages of felsic rocks are generally Archean andrange mainly between 25 and 39 Ga with two agepeaks at 26^28 Ga and 32^36 Ga (Wu et al 2005b)indicating Archean crust formation Petrologically theNCC is composed mainly of various gneisses granulitesamphibolites and supracrustal rocksThe NCC can be divided into Eastern and Western

blocks which collided along the Trans-North ChinaOrogen at 185 Ga (Zhao et al 2005) (Fig 1) TheWestern Block can be further divided into the YinshanBlock in the north and Ordos Block in the south whichare separated by the east^west-trending PaleoproterozoicKhondalite Belt (Zhao et al 2005) Within the EasternBlock a Paleoproterozoic collisional belt is also present onthe Liaodong^Jiaodong peninsulas (darker gray shadingin northeastern NCC in Fig 1 Faure et al 2004 Zhaoet al 2005 Lu et al 2006)From 185 Ga until the late Mesozoic the NCC

remained stable Starting in the Mesozoic intensive defor-mation mineralization and igneous activity occurredincluding extensive eruption of intermediate^acid volcanicrocks and widespread emplacement of granites (Wu et al2005a 2005c) During the Cenozoic the NCC has

witnessed extensive intraplate basaltic volcanism Someof these basalts contain abundant ultramafic xenoliths(Fig 1)In Shandong Province three stages of igneous rocks

from the Paleozoic Mesozoic and Cenozoic containmantle xenoliths and provide a unique opportunity tostudy the evolution of the lithospheric mantle throughoutthe critical period of lithosphere transformation within aspatially restricted region The Mengyin kimberliteemplaced during the Paleozoic at 470 Ma (Dobbs et al1994 Zhang amp Yang 2007 Yang et al 2009) containsabundant harzburgite xenoliths (Chi amp Lu 1996 Gaoet al 2002 Zheng et al 2007 Zhang et al 2008) Amongthe Mesozoic rocks the Tietonggou diorite in Laiwu(125 Ma) contains harzburgite and dunite xenolithssome of which yield Archean Os model ages (W L Xuet al 2008 Gao et al 2008) During the Cenozoic alkalinevolcanic activity occurred in numerous localities includingQixia (Gao et al 2002 Rudnick et al 2004) Penglai andShanwang (Fig 1) The Penglai volcanoes erupted duringthe Neogene at 57^42 Ma (Liu 1999) are located in thenortheastern portion of the Shandong Peninsula about500 km to the NE of the Mengyin kimberlite and about100 km to the north of Qixia The Shanwang volcanoeserupted at 16 Ma (Liu 1999 Zheng et al 2007) arelocated c 100 km to the north of the Paleozoic Mengyinkimberlite within theTan^Lu FaultMantle xenoliths in the Mengyin kimberlite are rounded

and average 5^8 cm across They have been intensivelyaltered and only garnet remains as a primary mineral(Chi amp Lu 1996 Zhang et al 2008 Gao et al 2002)In contrast lavas at both the Penglai and Shanwang volca-noes contain abundant fresh xenoliths of lherzolite andpyroxenite with minor harzburgite and wehrlitePhotographs of outcrop and hand specimens of somePenglai and Shanwang xenoliths are provided in an elec-tronic supplement (available for downloading at httpwwwpetrologyoxfordjournalsorg) Petrographic descrip-tions of the samples analysed in this study are given inAppendix AAt Penglai most xenoliths range in size between 5 and

10 cm These peridotite xenoliths are fresh generallycoarse-grained (1^3mm) and equigranular with someshowing porphyroclastic textures (Fig 2a and b) Nostrongly foliated types have been found Primary mineralsare olivine orthopyroxene (opx) clinopyroxene (cpx)and spinel (Fig 2a and b) most xenoliths are lherzolitesbut minor harzburgite also occurs (Table 1)The xenoliths at Shanwang are larger typically ranging

from 10 to 20 cm across These xenoliths are characterizedby four types of microstructures (protogranular transi-tional tabular equigranular and porphyroclastic texture)(Fig 2c and d) with primary mineral assemblages consist-ing of olivine cpx opx and spinel (Fig 2c and d) Again

CHU et al EASTERN NCC MANTLE THINING

1859

lherzolite dominates the rock types but harzburgites alsooccur as well as pyroxenites (SW04) and wehrlites (SW01SW21) (Table 1) The lack of garnet in peridotite xenolithsfrom both Penglai and Shanwang studied here indicatesderivation of these samples from depths of 580 kmReports of garnet-bearing lherzolites from Shanwanghowever indicate that some peridotites from this localewere derived from depths of 80 km (Zheng et al 2006)

ANALYT ICAL TECHNIQUESThe xenoliths were sawn from their lava hosts and the cutsurfaces were abraded with quartz in a sand blaster toremove any possible contamination from the saw bladeThe samples were first crushed using an alumina ceramicjaw crusher and then small chips devoid of surface alter-ation and host lava were ground into a fine powder usingan agate shatterbox A portion of the crushed fraction was

sieved and cpx separates were handpicked under a binocu-lar microscope to a purity of498

Elemental analysisMajor element data for whole-rock samples were obtainedby X-ray fluorescence spectrometry (XRF) on fused glassdisks using a Shimadzu XRF-1500 instrument at theInstitute of Geology and Geophysics Chinese Academyof Sciences (see Appendix B for details) Precisions are1^3 relative for elements present in concentrations41wt and about 10 relative for elements present inconcentrations510wt A Chinese ultramafic rock ref-erence material DZE-1 was analyzed during the sameperiod and the values determined are well within therange of consensus values (Appendix B Table B1)Trace element concentrations including the rare earth

elements (REE) were determined by inductively coupledplasma mass spectrometry (ICP-MS) using an Agilent

Fig 2 Photomicrographs of representative textures in mantle xenoliths from Penglai (a b) and Shanwang (c d) (a b) Penglai xenoliths PL18and PL19 show equigranular to porphyroclastic textures (plane-polarized light) (c) Shanwang xenolith SW16 with granuloblastic texture(plane-polarized light) (d) Shanwang xenolith SW18 with porphyroclastic texture (plane-polarized light) Petrographic descriptions of the sam-ples are given in Appendix A

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1860

Table 1 Major and trace element compositions of the studied xenoliths

MY9 MY10 MY33 MY34 MY35 PL01 PL02 PL03 PL06 PL07 PL08

Lithology Hz D Hz Hz Px Lz Hz Lz Lz Lz Lz

Modal compositions ()

Ol 589 752 531 561 560 732

Opx 257 205 301 248 259 193

Cpx 124 34 141 154 152 58

Sp 31 10 27 37 30 17

Fo 898 908 887 899 896 913

Major elements (wt )

SiO2 408 363 397 408 371 450 442 442 447 452 441

TiO2 010 004 049 005 032 011 004 022 018 013 006

Al2O3 032 006 027 079 867 356 119 397 431 396 169

TFe2O3 422 775 605 396 808 834 815 1092 828 847 797

MnO 007 007 005 005 013 011 012 014 011 012 011

MgO 394 446 387 393 289 393 445 361 382 379 438

CaO 064 021 044 046 272 261 101 279 327 308 139

Na2O 000 000 000 000 000 021 005 044 033 028 010

K2O 004 002 002 003 021 004 005 012 007 005 005

P2O5 005 003 003 002 020 002 002 007 004 002 003

LOI 140 105 140 140 134 017 013 055 003 013 010

Total 997 996 997 995 997 992 992 994 995 994 994

Trace elements (ppm)

Ni 2447 2729 2587 3227 427 2290 2743 2168 1964 1888 2704

Cu 174 652 387 252 342 167 139 281 242 237 258

Zn 238 381 233 336 527 578 511 102 611 578 495

Sr 401 239 328 503 225 203 199 132 523 245 374

Y 337 0264 388 298 393 318 0638 458 401 362 112

Zr 152 578 259 194 729 876 518 331 168 977 827

Nb 151 589 177 142 607 0640 130 754 327 121 234

Ba 242 661 338 142 308 732 765 580 230 130 107

Hf 0386 0113 161 113 142 0184 0085 0547 0407 0261 0151

Ta 0294 0259 0976 0322 246 0019 0048 0348 0155 0051 0093

Pb 0383 0699 0211 0229 0770 0330 0403 0856 113 127 0236

Th 0270 0564 0325 0155 758 0073 0075 0495 0311 0103 0157

U 0742 0088 0760 0495 171 0019 0024 0137 0092 0031 0047

S 480 510 210 440 140 550 550 260 50 550 50

Rare earth elements (ppm)

La 205 377 264 247 907 0464 0570 465 177 118 109

Ce 245 642 316 292 120 114 120 912 378 188 224

Pr 201 0629 241 229 100 0160 0137 127 0454 0318 0270

Nd 694 205 731 727 304 0828 0572 557 211 149 115

Sm 111 0262 100 0915 364 0280 0120 122 0580 0432 0252

Eu 0397 0076 0187 0233 0779 0103 0040 0420 0209 0157 0084

Gd 126 0206 114 0971 313 0399 0120 130 0723 0580 0258

Tb 0130 0018 0125 0095 0263 0071 0017 0175 0121 0100 0036

Dy 0635 0068 0648 0439 101 0478 0108 0946 0773 0677 0210

Ho 0109 0010 0130 0078 0152 0106 0021 0170 0167 0144 0040

Er 0262 0025 0338 0194 0386 0314 0061 0424 0477 0433 0114

Tm 0030 0003 0041 0024 0043 0044 0009 0053 0067 0062 0016

Yb 0167 0016 0235 0145 0274 0316 0070 0329 0442 0425 0114

Lu 0021 0002 0030 0021 0040 0047 0011 0048 0068 0066 0017

(continued)

CHU et al EASTERN NCC MANTLE THINING

1861

Table 1 Continued

PL10 PL11 PL12 PL13 PL14 PL15 PL16 PL17 PL18 PL19 PL23

Lithology Lz Lz Lz Lz Lz Lz Lz Lz Hz Lz Hz

Modal compositions ()

Ol 592 612 582 598 709 680 642

Opx 245 234 245 256 218 224 300

Cpx 128 125 144 113 47 76 37

Sp 34 29 29 33 27 20 21

Fo 895 904 881 903 907 904 911

Major elements (wt )

SiO2 444 443 442 446 446 450 440 447 437 445 456

TiO2 014 008 011 015 011 008 018 012 009 002 005

Al2O3 322 277 185 389 345 246 383 367 226 243 187

TFe2O3 838 830 815 852 797 845 1079 823 888 828 797

MnO 012 011 011 012 010 012 013 012 012 011 010

MgO 404 418 430 390 400 412 372 397 430 421 425

CaO 242 181 156 284 281 198 285 259 117 194 119

Na2O 027 012 018 024 024 016 034 022 009 007 007

K2O 006 007 014 004 004 005 005 006 006 003 004

P2O5 003 003 007 003 003 002 004 003 003 001 002

LOI 027 012 023 002 005 023 017 002 003 028 005

Total 991 992 992 994 993 992 995 994 994 992 993

Trace elements (ppm)

Ni 2356 2178 2232 2284 2039 2020 1871 2312 2260 2115 2113

Cu 209 148 115 176 152 112 253 246 160 122 120

Zn 617 618 519 593 574 464 796 595 603 520 532

Sr 514 438 113 436 398 341 624 499 468 184 250

Y 318 133 188 382 303 191 334 290 133 108 0634

Zr 128 117 165 140 107 813 196 121 114 389 675

Nb 232 385 517 208 189 141 261 315 334 0826 165

Ba 293 303 459 124 129 141 131 193 227 700 109

Hf 0286 0240 0289 0312 0271 0169 0389 0257 0254 0083 0159

Ta 0111 0199 0235 0095 0099 0066 0134 0147 0178 0043 0090

Pb 0748 0471 0671 0446 0685 0544 0347 0436 0417 0236 0595

Th 0180 0317 0459 0148 0182 0140 0245 0241 0264 0141 0184

U 0059 0097 0125 0054 0054 0038 0073 0070 0077 0046 0055

S 550 550 550 550 550 550 60 550 50 550 50

Rare earth elements (ppm)

La 152 218 380 122 139 0836 196 175 177 0873 0985

Ce 305 350 748 263 289 189 435 359 388 177 218

Pr 0387 0440 0842 0345 0348 0220 0539 0386 0402 0183 0226

Nd 171 177 335 162 156 0980 244 163 170 0721 0939

Sm 0447 0354 0671 0451 0410 0253 0602 0403 0367 0143 0187

Eu 0173 0119 0221 0174 0155 0099 0211 0144 0121 0049 0064

Gd 0539 0361 0641 0605 0527 0320 0682 0492 0382 0164 0188

Tb 0090 0049 0081 0100 0090 0053 0107 0080 0054 0029 0026

Dy 0570 0275 0416 0663 0571 0337 0643 0501 0303 0202 0133

Ho 0119 0055 0069 0139 0123 0071 0124 0106 0055 0044 0026

Er 0343 0148 0174 0405 0355 0218 0341 0298 0153 0142 0071

Tm 0048 0021 0021 0059 0049 0030 0047 0043 0020 0022 0011

Yb 0322 0151 0130 0381 0330 0216 0313 0299 0144 0163 0079

Lu 0050 0025 0019 0059 0051 0034 0044 0045 0023 0027 0014

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1862

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 2: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

kimberlites in Mengyin Shandong Province and FuxianLiaoning Province indicate the presence of lithospherewith a thickness of 200 km at the time of volcanism(Menzies et al 1993 Griffin et al 1998 Wang et al 1998Menzies amp Xu 1998) (Fig 1) Nevertheless geophysicaldata and petrological studies of mantle xenoliths collectedfrom Cenozoic basalts suggest that the present lithosphereis much thinner (60^120 km) and considerably more fertile(Menzies et al 1993 Griffin et al 1998 Menzies amp Xu1998 Fan et al 2000 Xu 2001 Rudnick et al 2004 Chenet al 2006) Because of these extensive studies the easternNCC is now recognized as the best example of a reacti-vated craton with decoupled crust and mantle lithosphere(Carlson et al 2005 Menzies et al 2007)Despite extensive investigation the cause of lithospheric

thinning beneath the eastern block of the NCC remainshighly debated One school of thought suggests that theoriginal Archean lithospheric mantle and associated lowercrust were removed via density foundering and werereplaced by peridotite residues formed by melting of con-vecting upper mantle during the Mesozoic to earlyCenozoic (Gao et al 2002 2004 2008 Wu et al 20032005a) Another model suggests that mantle with

Proterozoic Os model ages originated by mixing the origi-nal Archean mantle lithosphere with juvenile astheno-spheric mantle via melt^peridotite interaction (Zhang2005 Zhang et al 2008) Others have proposed that thismantle lithosphere might be exotic derived from southernChina or elsewhere and has no relationship with thecrustal^mantle evolution of the NCC (Wu et al 2006)Mapping of spatial variations in mantle melt depletion

ages of the lithospheric mantle underlying the NCCbefore and after the loss of the deep lithosphere is essentialto diagnose the cause of the loss (Carlson et al 2005Foley 2008) However despite much recent study the agestructure of the lithospheric mantle underlying the NCCremains unclear partly because of the paucity of data forappropriate xenolith locations This is especially true forthe easternmost portion of the NCC To date Os isotopicdata have been reported for peridotite xenoliths from thePaleozoic Mengyin Fuxian and Tieling kimberlites Thesedata reveal the existence of Archean lithospheric mantlebeneath the eastern NCC during the early Paleozoic (Gaoet al 2002Wu et al 2006 Zhang et al 2008) In contrastperidotite xenoliths from Cenozoic basalts in QixiaLonggang and Kuandian (Fig 1) show considerable

Fig 1 Map of the North China Craton showing the xenolith localities discussed in the text Tectonic subdivisions are based on Zhao et al(2005) Inset shows location of the North China Craton (NCC) relative to other cratonic blocks (eg South China Craton) and interveningfold belts

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1858

variations in Os model ages These model ages range fromProterozoic to Phanerozoic and have been interpreted indifferent ways Most importantly no peridotite xenolithswith Archean Os model ages have been documented inthe Cenozoic basalts from the eastern NCC (Gao et al2002Wu et al 2003 2006)With the intent of providing a much more detailed pic-

ture of the age structure of the mantle lithosphere underly-ing the eastern portion of the NCC we have determinedthe petrography major and trace element chemistry Sr^Nd^Hf^Os isotope and platinum-group element (PGE)concentrations for two large suites of peridotite xenolithsfrom the Cenozoic Penglai and Shanwang basalts inShandong Province These data complement publisheddata for the nearby Qixia peridotites (Gao et al 2002Rudnick et al 2004)To bolster limited earlier data and fur-ther constrain the nature of the lithospheric mantle presentduring the Paleozoic we also analyzed additional mantlexenoliths from the Paleozoic Mengyin kimberlites

GEOLOGICAL BACKGROUNDAND SAMPLE DESCR IPT IONSEastern China is composed of the three main lithotectonicdomains the Central Asian Orogenic Belt and the NCCin the north the Dabie^Sulu ultrahigh-pressure collisionalbelt in the central region and the South China Craton(including the Yangtze Craton and Southeastern Chinaorogenic belt) in the south (Fig 1) The NCC is the oldesttectonic unit in China containing crust as old as 3800Ma (Liu et al 1992 Wu et al 2008) Neodymium TDM

model ages of felsic rocks are generally Archean andrange mainly between 25 and 39 Ga with two agepeaks at 26^28 Ga and 32^36 Ga (Wu et al 2005b)indicating Archean crust formation Petrologically theNCC is composed mainly of various gneisses granulitesamphibolites and supracrustal rocksThe NCC can be divided into Eastern and Western

blocks which collided along the Trans-North ChinaOrogen at 185 Ga (Zhao et al 2005) (Fig 1) TheWestern Block can be further divided into the YinshanBlock in the north and Ordos Block in the south whichare separated by the east^west-trending PaleoproterozoicKhondalite Belt (Zhao et al 2005) Within the EasternBlock a Paleoproterozoic collisional belt is also present onthe Liaodong^Jiaodong peninsulas (darker gray shadingin northeastern NCC in Fig 1 Faure et al 2004 Zhaoet al 2005 Lu et al 2006)From 185 Ga until the late Mesozoic the NCC

remained stable Starting in the Mesozoic intensive defor-mation mineralization and igneous activity occurredincluding extensive eruption of intermediate^acid volcanicrocks and widespread emplacement of granites (Wu et al2005a 2005c) During the Cenozoic the NCC has

witnessed extensive intraplate basaltic volcanism Someof these basalts contain abundant ultramafic xenoliths(Fig 1)In Shandong Province three stages of igneous rocks

from the Paleozoic Mesozoic and Cenozoic containmantle xenoliths and provide a unique opportunity tostudy the evolution of the lithospheric mantle throughoutthe critical period of lithosphere transformation within aspatially restricted region The Mengyin kimberliteemplaced during the Paleozoic at 470 Ma (Dobbs et al1994 Zhang amp Yang 2007 Yang et al 2009) containsabundant harzburgite xenoliths (Chi amp Lu 1996 Gaoet al 2002 Zheng et al 2007 Zhang et al 2008) Amongthe Mesozoic rocks the Tietonggou diorite in Laiwu(125 Ma) contains harzburgite and dunite xenolithssome of which yield Archean Os model ages (W L Xuet al 2008 Gao et al 2008) During the Cenozoic alkalinevolcanic activity occurred in numerous localities includingQixia (Gao et al 2002 Rudnick et al 2004) Penglai andShanwang (Fig 1) The Penglai volcanoes erupted duringthe Neogene at 57^42 Ma (Liu 1999) are located in thenortheastern portion of the Shandong Peninsula about500 km to the NE of the Mengyin kimberlite and about100 km to the north of Qixia The Shanwang volcanoeserupted at 16 Ma (Liu 1999 Zheng et al 2007) arelocated c 100 km to the north of the Paleozoic Mengyinkimberlite within theTan^Lu FaultMantle xenoliths in the Mengyin kimberlite are rounded

and average 5^8 cm across They have been intensivelyaltered and only garnet remains as a primary mineral(Chi amp Lu 1996 Zhang et al 2008 Gao et al 2002)In contrast lavas at both the Penglai and Shanwang volca-noes contain abundant fresh xenoliths of lherzolite andpyroxenite with minor harzburgite and wehrlitePhotographs of outcrop and hand specimens of somePenglai and Shanwang xenoliths are provided in an elec-tronic supplement (available for downloading at httpwwwpetrologyoxfordjournalsorg) Petrographic descrip-tions of the samples analysed in this study are given inAppendix AAt Penglai most xenoliths range in size between 5 and

10 cm These peridotite xenoliths are fresh generallycoarse-grained (1^3mm) and equigranular with someshowing porphyroclastic textures (Fig 2a and b) Nostrongly foliated types have been found Primary mineralsare olivine orthopyroxene (opx) clinopyroxene (cpx)and spinel (Fig 2a and b) most xenoliths are lherzolitesbut minor harzburgite also occurs (Table 1)The xenoliths at Shanwang are larger typically ranging

from 10 to 20 cm across These xenoliths are characterizedby four types of microstructures (protogranular transi-tional tabular equigranular and porphyroclastic texture)(Fig 2c and d) with primary mineral assemblages consist-ing of olivine cpx opx and spinel (Fig 2c and d) Again

CHU et al EASTERN NCC MANTLE THINING

1859

lherzolite dominates the rock types but harzburgites alsooccur as well as pyroxenites (SW04) and wehrlites (SW01SW21) (Table 1) The lack of garnet in peridotite xenolithsfrom both Penglai and Shanwang studied here indicatesderivation of these samples from depths of 580 kmReports of garnet-bearing lherzolites from Shanwanghowever indicate that some peridotites from this localewere derived from depths of 80 km (Zheng et al 2006)

ANALYT ICAL TECHNIQUESThe xenoliths were sawn from their lava hosts and the cutsurfaces were abraded with quartz in a sand blaster toremove any possible contamination from the saw bladeThe samples were first crushed using an alumina ceramicjaw crusher and then small chips devoid of surface alter-ation and host lava were ground into a fine powder usingan agate shatterbox A portion of the crushed fraction was

sieved and cpx separates were handpicked under a binocu-lar microscope to a purity of498

Elemental analysisMajor element data for whole-rock samples were obtainedby X-ray fluorescence spectrometry (XRF) on fused glassdisks using a Shimadzu XRF-1500 instrument at theInstitute of Geology and Geophysics Chinese Academyof Sciences (see Appendix B for details) Precisions are1^3 relative for elements present in concentrations41wt and about 10 relative for elements present inconcentrations510wt A Chinese ultramafic rock ref-erence material DZE-1 was analyzed during the sameperiod and the values determined are well within therange of consensus values (Appendix B Table B1)Trace element concentrations including the rare earth

elements (REE) were determined by inductively coupledplasma mass spectrometry (ICP-MS) using an Agilent

Fig 2 Photomicrographs of representative textures in mantle xenoliths from Penglai (a b) and Shanwang (c d) (a b) Penglai xenoliths PL18and PL19 show equigranular to porphyroclastic textures (plane-polarized light) (c) Shanwang xenolith SW16 with granuloblastic texture(plane-polarized light) (d) Shanwang xenolith SW18 with porphyroclastic texture (plane-polarized light) Petrographic descriptions of the sam-ples are given in Appendix A

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1860

Table 1 Major and trace element compositions of the studied xenoliths

MY9 MY10 MY33 MY34 MY35 PL01 PL02 PL03 PL06 PL07 PL08

Lithology Hz D Hz Hz Px Lz Hz Lz Lz Lz Lz

Modal compositions ()

Ol 589 752 531 561 560 732

Opx 257 205 301 248 259 193

Cpx 124 34 141 154 152 58

Sp 31 10 27 37 30 17

Fo 898 908 887 899 896 913

Major elements (wt )

SiO2 408 363 397 408 371 450 442 442 447 452 441

TiO2 010 004 049 005 032 011 004 022 018 013 006

Al2O3 032 006 027 079 867 356 119 397 431 396 169

TFe2O3 422 775 605 396 808 834 815 1092 828 847 797

MnO 007 007 005 005 013 011 012 014 011 012 011

MgO 394 446 387 393 289 393 445 361 382 379 438

CaO 064 021 044 046 272 261 101 279 327 308 139

Na2O 000 000 000 000 000 021 005 044 033 028 010

K2O 004 002 002 003 021 004 005 012 007 005 005

P2O5 005 003 003 002 020 002 002 007 004 002 003

LOI 140 105 140 140 134 017 013 055 003 013 010

Total 997 996 997 995 997 992 992 994 995 994 994

Trace elements (ppm)

Ni 2447 2729 2587 3227 427 2290 2743 2168 1964 1888 2704

Cu 174 652 387 252 342 167 139 281 242 237 258

Zn 238 381 233 336 527 578 511 102 611 578 495

Sr 401 239 328 503 225 203 199 132 523 245 374

Y 337 0264 388 298 393 318 0638 458 401 362 112

Zr 152 578 259 194 729 876 518 331 168 977 827

Nb 151 589 177 142 607 0640 130 754 327 121 234

Ba 242 661 338 142 308 732 765 580 230 130 107

Hf 0386 0113 161 113 142 0184 0085 0547 0407 0261 0151

Ta 0294 0259 0976 0322 246 0019 0048 0348 0155 0051 0093

Pb 0383 0699 0211 0229 0770 0330 0403 0856 113 127 0236

Th 0270 0564 0325 0155 758 0073 0075 0495 0311 0103 0157

U 0742 0088 0760 0495 171 0019 0024 0137 0092 0031 0047

S 480 510 210 440 140 550 550 260 50 550 50

Rare earth elements (ppm)

La 205 377 264 247 907 0464 0570 465 177 118 109

Ce 245 642 316 292 120 114 120 912 378 188 224

Pr 201 0629 241 229 100 0160 0137 127 0454 0318 0270

Nd 694 205 731 727 304 0828 0572 557 211 149 115

Sm 111 0262 100 0915 364 0280 0120 122 0580 0432 0252

Eu 0397 0076 0187 0233 0779 0103 0040 0420 0209 0157 0084

Gd 126 0206 114 0971 313 0399 0120 130 0723 0580 0258

Tb 0130 0018 0125 0095 0263 0071 0017 0175 0121 0100 0036

Dy 0635 0068 0648 0439 101 0478 0108 0946 0773 0677 0210

Ho 0109 0010 0130 0078 0152 0106 0021 0170 0167 0144 0040

Er 0262 0025 0338 0194 0386 0314 0061 0424 0477 0433 0114

Tm 0030 0003 0041 0024 0043 0044 0009 0053 0067 0062 0016

Yb 0167 0016 0235 0145 0274 0316 0070 0329 0442 0425 0114

Lu 0021 0002 0030 0021 0040 0047 0011 0048 0068 0066 0017

(continued)

CHU et al EASTERN NCC MANTLE THINING

1861

Table 1 Continued

PL10 PL11 PL12 PL13 PL14 PL15 PL16 PL17 PL18 PL19 PL23

Lithology Lz Lz Lz Lz Lz Lz Lz Lz Hz Lz Hz

Modal compositions ()

Ol 592 612 582 598 709 680 642

Opx 245 234 245 256 218 224 300

Cpx 128 125 144 113 47 76 37

Sp 34 29 29 33 27 20 21

Fo 895 904 881 903 907 904 911

Major elements (wt )

SiO2 444 443 442 446 446 450 440 447 437 445 456

TiO2 014 008 011 015 011 008 018 012 009 002 005

Al2O3 322 277 185 389 345 246 383 367 226 243 187

TFe2O3 838 830 815 852 797 845 1079 823 888 828 797

MnO 012 011 011 012 010 012 013 012 012 011 010

MgO 404 418 430 390 400 412 372 397 430 421 425

CaO 242 181 156 284 281 198 285 259 117 194 119

Na2O 027 012 018 024 024 016 034 022 009 007 007

K2O 006 007 014 004 004 005 005 006 006 003 004

P2O5 003 003 007 003 003 002 004 003 003 001 002

LOI 027 012 023 002 005 023 017 002 003 028 005

Total 991 992 992 994 993 992 995 994 994 992 993

Trace elements (ppm)

Ni 2356 2178 2232 2284 2039 2020 1871 2312 2260 2115 2113

Cu 209 148 115 176 152 112 253 246 160 122 120

Zn 617 618 519 593 574 464 796 595 603 520 532

Sr 514 438 113 436 398 341 624 499 468 184 250

Y 318 133 188 382 303 191 334 290 133 108 0634

Zr 128 117 165 140 107 813 196 121 114 389 675

Nb 232 385 517 208 189 141 261 315 334 0826 165

Ba 293 303 459 124 129 141 131 193 227 700 109

Hf 0286 0240 0289 0312 0271 0169 0389 0257 0254 0083 0159

Ta 0111 0199 0235 0095 0099 0066 0134 0147 0178 0043 0090

Pb 0748 0471 0671 0446 0685 0544 0347 0436 0417 0236 0595

Th 0180 0317 0459 0148 0182 0140 0245 0241 0264 0141 0184

U 0059 0097 0125 0054 0054 0038 0073 0070 0077 0046 0055

S 550 550 550 550 550 550 60 550 50 550 50

Rare earth elements (ppm)

La 152 218 380 122 139 0836 196 175 177 0873 0985

Ce 305 350 748 263 289 189 435 359 388 177 218

Pr 0387 0440 0842 0345 0348 0220 0539 0386 0402 0183 0226

Nd 171 177 335 162 156 0980 244 163 170 0721 0939

Sm 0447 0354 0671 0451 0410 0253 0602 0403 0367 0143 0187

Eu 0173 0119 0221 0174 0155 0099 0211 0144 0121 0049 0064

Gd 0539 0361 0641 0605 0527 0320 0682 0492 0382 0164 0188

Tb 0090 0049 0081 0100 0090 0053 0107 0080 0054 0029 0026

Dy 0570 0275 0416 0663 0571 0337 0643 0501 0303 0202 0133

Ho 0119 0055 0069 0139 0123 0071 0124 0106 0055 0044 0026

Er 0343 0148 0174 0405 0355 0218 0341 0298 0153 0142 0071

Tm 0048 0021 0021 0059 0049 0030 0047 0043 0020 0022 0011

Yb 0322 0151 0130 0381 0330 0216 0313 0299 0144 0163 0079

Lu 0050 0025 0019 0059 0051 0034 0044 0045 0023 0027 0014

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1862

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 3: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

variations in Os model ages These model ages range fromProterozoic to Phanerozoic and have been interpreted indifferent ways Most importantly no peridotite xenolithswith Archean Os model ages have been documented inthe Cenozoic basalts from the eastern NCC (Gao et al2002Wu et al 2003 2006)With the intent of providing a much more detailed pic-

ture of the age structure of the mantle lithosphere underly-ing the eastern portion of the NCC we have determinedthe petrography major and trace element chemistry Sr^Nd^Hf^Os isotope and platinum-group element (PGE)concentrations for two large suites of peridotite xenolithsfrom the Cenozoic Penglai and Shanwang basalts inShandong Province These data complement publisheddata for the nearby Qixia peridotites (Gao et al 2002Rudnick et al 2004)To bolster limited earlier data and fur-ther constrain the nature of the lithospheric mantle presentduring the Paleozoic we also analyzed additional mantlexenoliths from the Paleozoic Mengyin kimberlites

GEOLOGICAL BACKGROUNDAND SAMPLE DESCR IPT IONSEastern China is composed of the three main lithotectonicdomains the Central Asian Orogenic Belt and the NCCin the north the Dabie^Sulu ultrahigh-pressure collisionalbelt in the central region and the South China Craton(including the Yangtze Craton and Southeastern Chinaorogenic belt) in the south (Fig 1) The NCC is the oldesttectonic unit in China containing crust as old as 3800Ma (Liu et al 1992 Wu et al 2008) Neodymium TDM

model ages of felsic rocks are generally Archean andrange mainly between 25 and 39 Ga with two agepeaks at 26^28 Ga and 32^36 Ga (Wu et al 2005b)indicating Archean crust formation Petrologically theNCC is composed mainly of various gneisses granulitesamphibolites and supracrustal rocksThe NCC can be divided into Eastern and Western

blocks which collided along the Trans-North ChinaOrogen at 185 Ga (Zhao et al 2005) (Fig 1) TheWestern Block can be further divided into the YinshanBlock in the north and Ordos Block in the south whichare separated by the east^west-trending PaleoproterozoicKhondalite Belt (Zhao et al 2005) Within the EasternBlock a Paleoproterozoic collisional belt is also present onthe Liaodong^Jiaodong peninsulas (darker gray shadingin northeastern NCC in Fig 1 Faure et al 2004 Zhaoet al 2005 Lu et al 2006)From 185 Ga until the late Mesozoic the NCC

remained stable Starting in the Mesozoic intensive defor-mation mineralization and igneous activity occurredincluding extensive eruption of intermediate^acid volcanicrocks and widespread emplacement of granites (Wu et al2005a 2005c) During the Cenozoic the NCC has

witnessed extensive intraplate basaltic volcanism Someof these basalts contain abundant ultramafic xenoliths(Fig 1)In Shandong Province three stages of igneous rocks

from the Paleozoic Mesozoic and Cenozoic containmantle xenoliths and provide a unique opportunity tostudy the evolution of the lithospheric mantle throughoutthe critical period of lithosphere transformation within aspatially restricted region The Mengyin kimberliteemplaced during the Paleozoic at 470 Ma (Dobbs et al1994 Zhang amp Yang 2007 Yang et al 2009) containsabundant harzburgite xenoliths (Chi amp Lu 1996 Gaoet al 2002 Zheng et al 2007 Zhang et al 2008) Amongthe Mesozoic rocks the Tietonggou diorite in Laiwu(125 Ma) contains harzburgite and dunite xenolithssome of which yield Archean Os model ages (W L Xuet al 2008 Gao et al 2008) During the Cenozoic alkalinevolcanic activity occurred in numerous localities includingQixia (Gao et al 2002 Rudnick et al 2004) Penglai andShanwang (Fig 1) The Penglai volcanoes erupted duringthe Neogene at 57^42 Ma (Liu 1999) are located in thenortheastern portion of the Shandong Peninsula about500 km to the NE of the Mengyin kimberlite and about100 km to the north of Qixia The Shanwang volcanoeserupted at 16 Ma (Liu 1999 Zheng et al 2007) arelocated c 100 km to the north of the Paleozoic Mengyinkimberlite within theTan^Lu FaultMantle xenoliths in the Mengyin kimberlite are rounded

and average 5^8 cm across They have been intensivelyaltered and only garnet remains as a primary mineral(Chi amp Lu 1996 Zhang et al 2008 Gao et al 2002)In contrast lavas at both the Penglai and Shanwang volca-noes contain abundant fresh xenoliths of lherzolite andpyroxenite with minor harzburgite and wehrlitePhotographs of outcrop and hand specimens of somePenglai and Shanwang xenoliths are provided in an elec-tronic supplement (available for downloading at httpwwwpetrologyoxfordjournalsorg) Petrographic descrip-tions of the samples analysed in this study are given inAppendix AAt Penglai most xenoliths range in size between 5 and

10 cm These peridotite xenoliths are fresh generallycoarse-grained (1^3mm) and equigranular with someshowing porphyroclastic textures (Fig 2a and b) Nostrongly foliated types have been found Primary mineralsare olivine orthopyroxene (opx) clinopyroxene (cpx)and spinel (Fig 2a and b) most xenoliths are lherzolitesbut minor harzburgite also occurs (Table 1)The xenoliths at Shanwang are larger typically ranging

from 10 to 20 cm across These xenoliths are characterizedby four types of microstructures (protogranular transi-tional tabular equigranular and porphyroclastic texture)(Fig 2c and d) with primary mineral assemblages consist-ing of olivine cpx opx and spinel (Fig 2c and d) Again

CHU et al EASTERN NCC MANTLE THINING

1859

lherzolite dominates the rock types but harzburgites alsooccur as well as pyroxenites (SW04) and wehrlites (SW01SW21) (Table 1) The lack of garnet in peridotite xenolithsfrom both Penglai and Shanwang studied here indicatesderivation of these samples from depths of 580 kmReports of garnet-bearing lherzolites from Shanwanghowever indicate that some peridotites from this localewere derived from depths of 80 km (Zheng et al 2006)

ANALYT ICAL TECHNIQUESThe xenoliths were sawn from their lava hosts and the cutsurfaces were abraded with quartz in a sand blaster toremove any possible contamination from the saw bladeThe samples were first crushed using an alumina ceramicjaw crusher and then small chips devoid of surface alter-ation and host lava were ground into a fine powder usingan agate shatterbox A portion of the crushed fraction was

sieved and cpx separates were handpicked under a binocu-lar microscope to a purity of498

Elemental analysisMajor element data for whole-rock samples were obtainedby X-ray fluorescence spectrometry (XRF) on fused glassdisks using a Shimadzu XRF-1500 instrument at theInstitute of Geology and Geophysics Chinese Academyof Sciences (see Appendix B for details) Precisions are1^3 relative for elements present in concentrations41wt and about 10 relative for elements present inconcentrations510wt A Chinese ultramafic rock ref-erence material DZE-1 was analyzed during the sameperiod and the values determined are well within therange of consensus values (Appendix B Table B1)Trace element concentrations including the rare earth

elements (REE) were determined by inductively coupledplasma mass spectrometry (ICP-MS) using an Agilent

Fig 2 Photomicrographs of representative textures in mantle xenoliths from Penglai (a b) and Shanwang (c d) (a b) Penglai xenoliths PL18and PL19 show equigranular to porphyroclastic textures (plane-polarized light) (c) Shanwang xenolith SW16 with granuloblastic texture(plane-polarized light) (d) Shanwang xenolith SW18 with porphyroclastic texture (plane-polarized light) Petrographic descriptions of the sam-ples are given in Appendix A

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1860

Table 1 Major and trace element compositions of the studied xenoliths

MY9 MY10 MY33 MY34 MY35 PL01 PL02 PL03 PL06 PL07 PL08

Lithology Hz D Hz Hz Px Lz Hz Lz Lz Lz Lz

Modal compositions ()

Ol 589 752 531 561 560 732

Opx 257 205 301 248 259 193

Cpx 124 34 141 154 152 58

Sp 31 10 27 37 30 17

Fo 898 908 887 899 896 913

Major elements (wt )

SiO2 408 363 397 408 371 450 442 442 447 452 441

TiO2 010 004 049 005 032 011 004 022 018 013 006

Al2O3 032 006 027 079 867 356 119 397 431 396 169

TFe2O3 422 775 605 396 808 834 815 1092 828 847 797

MnO 007 007 005 005 013 011 012 014 011 012 011

MgO 394 446 387 393 289 393 445 361 382 379 438

CaO 064 021 044 046 272 261 101 279 327 308 139

Na2O 000 000 000 000 000 021 005 044 033 028 010

K2O 004 002 002 003 021 004 005 012 007 005 005

P2O5 005 003 003 002 020 002 002 007 004 002 003

LOI 140 105 140 140 134 017 013 055 003 013 010

Total 997 996 997 995 997 992 992 994 995 994 994

Trace elements (ppm)

Ni 2447 2729 2587 3227 427 2290 2743 2168 1964 1888 2704

Cu 174 652 387 252 342 167 139 281 242 237 258

Zn 238 381 233 336 527 578 511 102 611 578 495

Sr 401 239 328 503 225 203 199 132 523 245 374

Y 337 0264 388 298 393 318 0638 458 401 362 112

Zr 152 578 259 194 729 876 518 331 168 977 827

Nb 151 589 177 142 607 0640 130 754 327 121 234

Ba 242 661 338 142 308 732 765 580 230 130 107

Hf 0386 0113 161 113 142 0184 0085 0547 0407 0261 0151

Ta 0294 0259 0976 0322 246 0019 0048 0348 0155 0051 0093

Pb 0383 0699 0211 0229 0770 0330 0403 0856 113 127 0236

Th 0270 0564 0325 0155 758 0073 0075 0495 0311 0103 0157

U 0742 0088 0760 0495 171 0019 0024 0137 0092 0031 0047

S 480 510 210 440 140 550 550 260 50 550 50

Rare earth elements (ppm)

La 205 377 264 247 907 0464 0570 465 177 118 109

Ce 245 642 316 292 120 114 120 912 378 188 224

Pr 201 0629 241 229 100 0160 0137 127 0454 0318 0270

Nd 694 205 731 727 304 0828 0572 557 211 149 115

Sm 111 0262 100 0915 364 0280 0120 122 0580 0432 0252

Eu 0397 0076 0187 0233 0779 0103 0040 0420 0209 0157 0084

Gd 126 0206 114 0971 313 0399 0120 130 0723 0580 0258

Tb 0130 0018 0125 0095 0263 0071 0017 0175 0121 0100 0036

Dy 0635 0068 0648 0439 101 0478 0108 0946 0773 0677 0210

Ho 0109 0010 0130 0078 0152 0106 0021 0170 0167 0144 0040

Er 0262 0025 0338 0194 0386 0314 0061 0424 0477 0433 0114

Tm 0030 0003 0041 0024 0043 0044 0009 0053 0067 0062 0016

Yb 0167 0016 0235 0145 0274 0316 0070 0329 0442 0425 0114

Lu 0021 0002 0030 0021 0040 0047 0011 0048 0068 0066 0017

(continued)

CHU et al EASTERN NCC MANTLE THINING

1861

Table 1 Continued

PL10 PL11 PL12 PL13 PL14 PL15 PL16 PL17 PL18 PL19 PL23

Lithology Lz Lz Lz Lz Lz Lz Lz Lz Hz Lz Hz

Modal compositions ()

Ol 592 612 582 598 709 680 642

Opx 245 234 245 256 218 224 300

Cpx 128 125 144 113 47 76 37

Sp 34 29 29 33 27 20 21

Fo 895 904 881 903 907 904 911

Major elements (wt )

SiO2 444 443 442 446 446 450 440 447 437 445 456

TiO2 014 008 011 015 011 008 018 012 009 002 005

Al2O3 322 277 185 389 345 246 383 367 226 243 187

TFe2O3 838 830 815 852 797 845 1079 823 888 828 797

MnO 012 011 011 012 010 012 013 012 012 011 010

MgO 404 418 430 390 400 412 372 397 430 421 425

CaO 242 181 156 284 281 198 285 259 117 194 119

Na2O 027 012 018 024 024 016 034 022 009 007 007

K2O 006 007 014 004 004 005 005 006 006 003 004

P2O5 003 003 007 003 003 002 004 003 003 001 002

LOI 027 012 023 002 005 023 017 002 003 028 005

Total 991 992 992 994 993 992 995 994 994 992 993

Trace elements (ppm)

Ni 2356 2178 2232 2284 2039 2020 1871 2312 2260 2115 2113

Cu 209 148 115 176 152 112 253 246 160 122 120

Zn 617 618 519 593 574 464 796 595 603 520 532

Sr 514 438 113 436 398 341 624 499 468 184 250

Y 318 133 188 382 303 191 334 290 133 108 0634

Zr 128 117 165 140 107 813 196 121 114 389 675

Nb 232 385 517 208 189 141 261 315 334 0826 165

Ba 293 303 459 124 129 141 131 193 227 700 109

Hf 0286 0240 0289 0312 0271 0169 0389 0257 0254 0083 0159

Ta 0111 0199 0235 0095 0099 0066 0134 0147 0178 0043 0090

Pb 0748 0471 0671 0446 0685 0544 0347 0436 0417 0236 0595

Th 0180 0317 0459 0148 0182 0140 0245 0241 0264 0141 0184

U 0059 0097 0125 0054 0054 0038 0073 0070 0077 0046 0055

S 550 550 550 550 550 550 60 550 50 550 50

Rare earth elements (ppm)

La 152 218 380 122 139 0836 196 175 177 0873 0985

Ce 305 350 748 263 289 189 435 359 388 177 218

Pr 0387 0440 0842 0345 0348 0220 0539 0386 0402 0183 0226

Nd 171 177 335 162 156 0980 244 163 170 0721 0939

Sm 0447 0354 0671 0451 0410 0253 0602 0403 0367 0143 0187

Eu 0173 0119 0221 0174 0155 0099 0211 0144 0121 0049 0064

Gd 0539 0361 0641 0605 0527 0320 0682 0492 0382 0164 0188

Tb 0090 0049 0081 0100 0090 0053 0107 0080 0054 0029 0026

Dy 0570 0275 0416 0663 0571 0337 0643 0501 0303 0202 0133

Ho 0119 0055 0069 0139 0123 0071 0124 0106 0055 0044 0026

Er 0343 0148 0174 0405 0355 0218 0341 0298 0153 0142 0071

Tm 0048 0021 0021 0059 0049 0030 0047 0043 0020 0022 0011

Yb 0322 0151 0130 0381 0330 0216 0313 0299 0144 0163 0079

Lu 0050 0025 0019 0059 0051 0034 0044 0045 0023 0027 0014

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1862

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 4: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

lherzolite dominates the rock types but harzburgites alsooccur as well as pyroxenites (SW04) and wehrlites (SW01SW21) (Table 1) The lack of garnet in peridotite xenolithsfrom both Penglai and Shanwang studied here indicatesderivation of these samples from depths of 580 kmReports of garnet-bearing lherzolites from Shanwanghowever indicate that some peridotites from this localewere derived from depths of 80 km (Zheng et al 2006)

ANALYT ICAL TECHNIQUESThe xenoliths were sawn from their lava hosts and the cutsurfaces were abraded with quartz in a sand blaster toremove any possible contamination from the saw bladeThe samples were first crushed using an alumina ceramicjaw crusher and then small chips devoid of surface alter-ation and host lava were ground into a fine powder usingan agate shatterbox A portion of the crushed fraction was

sieved and cpx separates were handpicked under a binocu-lar microscope to a purity of498

Elemental analysisMajor element data for whole-rock samples were obtainedby X-ray fluorescence spectrometry (XRF) on fused glassdisks using a Shimadzu XRF-1500 instrument at theInstitute of Geology and Geophysics Chinese Academyof Sciences (see Appendix B for details) Precisions are1^3 relative for elements present in concentrations41wt and about 10 relative for elements present inconcentrations510wt A Chinese ultramafic rock ref-erence material DZE-1 was analyzed during the sameperiod and the values determined are well within therange of consensus values (Appendix B Table B1)Trace element concentrations including the rare earth

elements (REE) were determined by inductively coupledplasma mass spectrometry (ICP-MS) using an Agilent

Fig 2 Photomicrographs of representative textures in mantle xenoliths from Penglai (a b) and Shanwang (c d) (a b) Penglai xenoliths PL18and PL19 show equigranular to porphyroclastic textures (plane-polarized light) (c) Shanwang xenolith SW16 with granuloblastic texture(plane-polarized light) (d) Shanwang xenolith SW18 with porphyroclastic texture (plane-polarized light) Petrographic descriptions of the sam-ples are given in Appendix A

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1860

Table 1 Major and trace element compositions of the studied xenoliths

MY9 MY10 MY33 MY34 MY35 PL01 PL02 PL03 PL06 PL07 PL08

Lithology Hz D Hz Hz Px Lz Hz Lz Lz Lz Lz

Modal compositions ()

Ol 589 752 531 561 560 732

Opx 257 205 301 248 259 193

Cpx 124 34 141 154 152 58

Sp 31 10 27 37 30 17

Fo 898 908 887 899 896 913

Major elements (wt )

SiO2 408 363 397 408 371 450 442 442 447 452 441

TiO2 010 004 049 005 032 011 004 022 018 013 006

Al2O3 032 006 027 079 867 356 119 397 431 396 169

TFe2O3 422 775 605 396 808 834 815 1092 828 847 797

MnO 007 007 005 005 013 011 012 014 011 012 011

MgO 394 446 387 393 289 393 445 361 382 379 438

CaO 064 021 044 046 272 261 101 279 327 308 139

Na2O 000 000 000 000 000 021 005 044 033 028 010

K2O 004 002 002 003 021 004 005 012 007 005 005

P2O5 005 003 003 002 020 002 002 007 004 002 003

LOI 140 105 140 140 134 017 013 055 003 013 010

Total 997 996 997 995 997 992 992 994 995 994 994

Trace elements (ppm)

Ni 2447 2729 2587 3227 427 2290 2743 2168 1964 1888 2704

Cu 174 652 387 252 342 167 139 281 242 237 258

Zn 238 381 233 336 527 578 511 102 611 578 495

Sr 401 239 328 503 225 203 199 132 523 245 374

Y 337 0264 388 298 393 318 0638 458 401 362 112

Zr 152 578 259 194 729 876 518 331 168 977 827

Nb 151 589 177 142 607 0640 130 754 327 121 234

Ba 242 661 338 142 308 732 765 580 230 130 107

Hf 0386 0113 161 113 142 0184 0085 0547 0407 0261 0151

Ta 0294 0259 0976 0322 246 0019 0048 0348 0155 0051 0093

Pb 0383 0699 0211 0229 0770 0330 0403 0856 113 127 0236

Th 0270 0564 0325 0155 758 0073 0075 0495 0311 0103 0157

U 0742 0088 0760 0495 171 0019 0024 0137 0092 0031 0047

S 480 510 210 440 140 550 550 260 50 550 50

Rare earth elements (ppm)

La 205 377 264 247 907 0464 0570 465 177 118 109

Ce 245 642 316 292 120 114 120 912 378 188 224

Pr 201 0629 241 229 100 0160 0137 127 0454 0318 0270

Nd 694 205 731 727 304 0828 0572 557 211 149 115

Sm 111 0262 100 0915 364 0280 0120 122 0580 0432 0252

Eu 0397 0076 0187 0233 0779 0103 0040 0420 0209 0157 0084

Gd 126 0206 114 0971 313 0399 0120 130 0723 0580 0258

Tb 0130 0018 0125 0095 0263 0071 0017 0175 0121 0100 0036

Dy 0635 0068 0648 0439 101 0478 0108 0946 0773 0677 0210

Ho 0109 0010 0130 0078 0152 0106 0021 0170 0167 0144 0040

Er 0262 0025 0338 0194 0386 0314 0061 0424 0477 0433 0114

Tm 0030 0003 0041 0024 0043 0044 0009 0053 0067 0062 0016

Yb 0167 0016 0235 0145 0274 0316 0070 0329 0442 0425 0114

Lu 0021 0002 0030 0021 0040 0047 0011 0048 0068 0066 0017

(continued)

CHU et al EASTERN NCC MANTLE THINING

1861

Table 1 Continued

PL10 PL11 PL12 PL13 PL14 PL15 PL16 PL17 PL18 PL19 PL23

Lithology Lz Lz Lz Lz Lz Lz Lz Lz Hz Lz Hz

Modal compositions ()

Ol 592 612 582 598 709 680 642

Opx 245 234 245 256 218 224 300

Cpx 128 125 144 113 47 76 37

Sp 34 29 29 33 27 20 21

Fo 895 904 881 903 907 904 911

Major elements (wt )

SiO2 444 443 442 446 446 450 440 447 437 445 456

TiO2 014 008 011 015 011 008 018 012 009 002 005

Al2O3 322 277 185 389 345 246 383 367 226 243 187

TFe2O3 838 830 815 852 797 845 1079 823 888 828 797

MnO 012 011 011 012 010 012 013 012 012 011 010

MgO 404 418 430 390 400 412 372 397 430 421 425

CaO 242 181 156 284 281 198 285 259 117 194 119

Na2O 027 012 018 024 024 016 034 022 009 007 007

K2O 006 007 014 004 004 005 005 006 006 003 004

P2O5 003 003 007 003 003 002 004 003 003 001 002

LOI 027 012 023 002 005 023 017 002 003 028 005

Total 991 992 992 994 993 992 995 994 994 992 993

Trace elements (ppm)

Ni 2356 2178 2232 2284 2039 2020 1871 2312 2260 2115 2113

Cu 209 148 115 176 152 112 253 246 160 122 120

Zn 617 618 519 593 574 464 796 595 603 520 532

Sr 514 438 113 436 398 341 624 499 468 184 250

Y 318 133 188 382 303 191 334 290 133 108 0634

Zr 128 117 165 140 107 813 196 121 114 389 675

Nb 232 385 517 208 189 141 261 315 334 0826 165

Ba 293 303 459 124 129 141 131 193 227 700 109

Hf 0286 0240 0289 0312 0271 0169 0389 0257 0254 0083 0159

Ta 0111 0199 0235 0095 0099 0066 0134 0147 0178 0043 0090

Pb 0748 0471 0671 0446 0685 0544 0347 0436 0417 0236 0595

Th 0180 0317 0459 0148 0182 0140 0245 0241 0264 0141 0184

U 0059 0097 0125 0054 0054 0038 0073 0070 0077 0046 0055

S 550 550 550 550 550 550 60 550 50 550 50

Rare earth elements (ppm)

La 152 218 380 122 139 0836 196 175 177 0873 0985

Ce 305 350 748 263 289 189 435 359 388 177 218

Pr 0387 0440 0842 0345 0348 0220 0539 0386 0402 0183 0226

Nd 171 177 335 162 156 0980 244 163 170 0721 0939

Sm 0447 0354 0671 0451 0410 0253 0602 0403 0367 0143 0187

Eu 0173 0119 0221 0174 0155 0099 0211 0144 0121 0049 0064

Gd 0539 0361 0641 0605 0527 0320 0682 0492 0382 0164 0188

Tb 0090 0049 0081 0100 0090 0053 0107 0080 0054 0029 0026

Dy 0570 0275 0416 0663 0571 0337 0643 0501 0303 0202 0133

Ho 0119 0055 0069 0139 0123 0071 0124 0106 0055 0044 0026

Er 0343 0148 0174 0405 0355 0218 0341 0298 0153 0142 0071

Tm 0048 0021 0021 0059 0049 0030 0047 0043 0020 0022 0011

Yb 0322 0151 0130 0381 0330 0216 0313 0299 0144 0163 0079

Lu 0050 0025 0019 0059 0051 0034 0044 0045 0023 0027 0014

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1862

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 5: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 1 Major and trace element compositions of the studied xenoliths

MY9 MY10 MY33 MY34 MY35 PL01 PL02 PL03 PL06 PL07 PL08

Lithology Hz D Hz Hz Px Lz Hz Lz Lz Lz Lz

Modal compositions ()

Ol 589 752 531 561 560 732

Opx 257 205 301 248 259 193

Cpx 124 34 141 154 152 58

Sp 31 10 27 37 30 17

Fo 898 908 887 899 896 913

Major elements (wt )

SiO2 408 363 397 408 371 450 442 442 447 452 441

TiO2 010 004 049 005 032 011 004 022 018 013 006

Al2O3 032 006 027 079 867 356 119 397 431 396 169

TFe2O3 422 775 605 396 808 834 815 1092 828 847 797

MnO 007 007 005 005 013 011 012 014 011 012 011

MgO 394 446 387 393 289 393 445 361 382 379 438

CaO 064 021 044 046 272 261 101 279 327 308 139

Na2O 000 000 000 000 000 021 005 044 033 028 010

K2O 004 002 002 003 021 004 005 012 007 005 005

P2O5 005 003 003 002 020 002 002 007 004 002 003

LOI 140 105 140 140 134 017 013 055 003 013 010

Total 997 996 997 995 997 992 992 994 995 994 994

Trace elements (ppm)

Ni 2447 2729 2587 3227 427 2290 2743 2168 1964 1888 2704

Cu 174 652 387 252 342 167 139 281 242 237 258

Zn 238 381 233 336 527 578 511 102 611 578 495

Sr 401 239 328 503 225 203 199 132 523 245 374

Y 337 0264 388 298 393 318 0638 458 401 362 112

Zr 152 578 259 194 729 876 518 331 168 977 827

Nb 151 589 177 142 607 0640 130 754 327 121 234

Ba 242 661 338 142 308 732 765 580 230 130 107

Hf 0386 0113 161 113 142 0184 0085 0547 0407 0261 0151

Ta 0294 0259 0976 0322 246 0019 0048 0348 0155 0051 0093

Pb 0383 0699 0211 0229 0770 0330 0403 0856 113 127 0236

Th 0270 0564 0325 0155 758 0073 0075 0495 0311 0103 0157

U 0742 0088 0760 0495 171 0019 0024 0137 0092 0031 0047

S 480 510 210 440 140 550 550 260 50 550 50

Rare earth elements (ppm)

La 205 377 264 247 907 0464 0570 465 177 118 109

Ce 245 642 316 292 120 114 120 912 378 188 224

Pr 201 0629 241 229 100 0160 0137 127 0454 0318 0270

Nd 694 205 731 727 304 0828 0572 557 211 149 115

Sm 111 0262 100 0915 364 0280 0120 122 0580 0432 0252

Eu 0397 0076 0187 0233 0779 0103 0040 0420 0209 0157 0084

Gd 126 0206 114 0971 313 0399 0120 130 0723 0580 0258

Tb 0130 0018 0125 0095 0263 0071 0017 0175 0121 0100 0036

Dy 0635 0068 0648 0439 101 0478 0108 0946 0773 0677 0210

Ho 0109 0010 0130 0078 0152 0106 0021 0170 0167 0144 0040

Er 0262 0025 0338 0194 0386 0314 0061 0424 0477 0433 0114

Tm 0030 0003 0041 0024 0043 0044 0009 0053 0067 0062 0016

Yb 0167 0016 0235 0145 0274 0316 0070 0329 0442 0425 0114

Lu 0021 0002 0030 0021 0040 0047 0011 0048 0068 0066 0017

(continued)

CHU et al EASTERN NCC MANTLE THINING

1861

Table 1 Continued

PL10 PL11 PL12 PL13 PL14 PL15 PL16 PL17 PL18 PL19 PL23

Lithology Lz Lz Lz Lz Lz Lz Lz Lz Hz Lz Hz

Modal compositions ()

Ol 592 612 582 598 709 680 642

Opx 245 234 245 256 218 224 300

Cpx 128 125 144 113 47 76 37

Sp 34 29 29 33 27 20 21

Fo 895 904 881 903 907 904 911

Major elements (wt )

SiO2 444 443 442 446 446 450 440 447 437 445 456

TiO2 014 008 011 015 011 008 018 012 009 002 005

Al2O3 322 277 185 389 345 246 383 367 226 243 187

TFe2O3 838 830 815 852 797 845 1079 823 888 828 797

MnO 012 011 011 012 010 012 013 012 012 011 010

MgO 404 418 430 390 400 412 372 397 430 421 425

CaO 242 181 156 284 281 198 285 259 117 194 119

Na2O 027 012 018 024 024 016 034 022 009 007 007

K2O 006 007 014 004 004 005 005 006 006 003 004

P2O5 003 003 007 003 003 002 004 003 003 001 002

LOI 027 012 023 002 005 023 017 002 003 028 005

Total 991 992 992 994 993 992 995 994 994 992 993

Trace elements (ppm)

Ni 2356 2178 2232 2284 2039 2020 1871 2312 2260 2115 2113

Cu 209 148 115 176 152 112 253 246 160 122 120

Zn 617 618 519 593 574 464 796 595 603 520 532

Sr 514 438 113 436 398 341 624 499 468 184 250

Y 318 133 188 382 303 191 334 290 133 108 0634

Zr 128 117 165 140 107 813 196 121 114 389 675

Nb 232 385 517 208 189 141 261 315 334 0826 165

Ba 293 303 459 124 129 141 131 193 227 700 109

Hf 0286 0240 0289 0312 0271 0169 0389 0257 0254 0083 0159

Ta 0111 0199 0235 0095 0099 0066 0134 0147 0178 0043 0090

Pb 0748 0471 0671 0446 0685 0544 0347 0436 0417 0236 0595

Th 0180 0317 0459 0148 0182 0140 0245 0241 0264 0141 0184

U 0059 0097 0125 0054 0054 0038 0073 0070 0077 0046 0055

S 550 550 550 550 550 550 60 550 50 550 50

Rare earth elements (ppm)

La 152 218 380 122 139 0836 196 175 177 0873 0985

Ce 305 350 748 263 289 189 435 359 388 177 218

Pr 0387 0440 0842 0345 0348 0220 0539 0386 0402 0183 0226

Nd 171 177 335 162 156 0980 244 163 170 0721 0939

Sm 0447 0354 0671 0451 0410 0253 0602 0403 0367 0143 0187

Eu 0173 0119 0221 0174 0155 0099 0211 0144 0121 0049 0064

Gd 0539 0361 0641 0605 0527 0320 0682 0492 0382 0164 0188

Tb 0090 0049 0081 0100 0090 0053 0107 0080 0054 0029 0026

Dy 0570 0275 0416 0663 0571 0337 0643 0501 0303 0202 0133

Ho 0119 0055 0069 0139 0123 0071 0124 0106 0055 0044 0026

Er 0343 0148 0174 0405 0355 0218 0341 0298 0153 0142 0071

Tm 0048 0021 0021 0059 0049 0030 0047 0043 0020 0022 0011

Yb 0322 0151 0130 0381 0330 0216 0313 0299 0144 0163 0079

Lu 0050 0025 0019 0059 0051 0034 0044 0045 0023 0027 0014

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1862

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 6: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 1 Continued

PL10 PL11 PL12 PL13 PL14 PL15 PL16 PL17 PL18 PL19 PL23

Lithology Lz Lz Lz Lz Lz Lz Lz Lz Hz Lz Hz

Modal compositions ()

Ol 592 612 582 598 709 680 642

Opx 245 234 245 256 218 224 300

Cpx 128 125 144 113 47 76 37

Sp 34 29 29 33 27 20 21

Fo 895 904 881 903 907 904 911

Major elements (wt )

SiO2 444 443 442 446 446 450 440 447 437 445 456

TiO2 014 008 011 015 011 008 018 012 009 002 005

Al2O3 322 277 185 389 345 246 383 367 226 243 187

TFe2O3 838 830 815 852 797 845 1079 823 888 828 797

MnO 012 011 011 012 010 012 013 012 012 011 010

MgO 404 418 430 390 400 412 372 397 430 421 425

CaO 242 181 156 284 281 198 285 259 117 194 119

Na2O 027 012 018 024 024 016 034 022 009 007 007

K2O 006 007 014 004 004 005 005 006 006 003 004

P2O5 003 003 007 003 003 002 004 003 003 001 002

LOI 027 012 023 002 005 023 017 002 003 028 005

Total 991 992 992 994 993 992 995 994 994 992 993

Trace elements (ppm)

Ni 2356 2178 2232 2284 2039 2020 1871 2312 2260 2115 2113

Cu 209 148 115 176 152 112 253 246 160 122 120

Zn 617 618 519 593 574 464 796 595 603 520 532

Sr 514 438 113 436 398 341 624 499 468 184 250

Y 318 133 188 382 303 191 334 290 133 108 0634

Zr 128 117 165 140 107 813 196 121 114 389 675

Nb 232 385 517 208 189 141 261 315 334 0826 165

Ba 293 303 459 124 129 141 131 193 227 700 109

Hf 0286 0240 0289 0312 0271 0169 0389 0257 0254 0083 0159

Ta 0111 0199 0235 0095 0099 0066 0134 0147 0178 0043 0090

Pb 0748 0471 0671 0446 0685 0544 0347 0436 0417 0236 0595

Th 0180 0317 0459 0148 0182 0140 0245 0241 0264 0141 0184

U 0059 0097 0125 0054 0054 0038 0073 0070 0077 0046 0055

S 550 550 550 550 550 550 60 550 50 550 50

Rare earth elements (ppm)

La 152 218 380 122 139 0836 196 175 177 0873 0985

Ce 305 350 748 263 289 189 435 359 388 177 218

Pr 0387 0440 0842 0345 0348 0220 0539 0386 0402 0183 0226

Nd 171 177 335 162 156 0980 244 163 170 0721 0939

Sm 0447 0354 0671 0451 0410 0253 0602 0403 0367 0143 0187

Eu 0173 0119 0221 0174 0155 0099 0211 0144 0121 0049 0064

Gd 0539 0361 0641 0605 0527 0320 0682 0492 0382 0164 0188

Tb 0090 0049 0081 0100 0090 0053 0107 0080 0054 0029 0026

Dy 0570 0275 0416 0663 0571 0337 0643 0501 0303 0202 0133

Ho 0119 0055 0069 0139 0123 0071 0124 0106 0055 0044 0026

Er 0343 0148 0174 0405 0355 0218 0341 0298 0153 0142 0071

Tm 0048 0021 0021 0059 0049 0030 0047 0043 0020 0022 0011

Yb 0322 0151 0130 0381 0330 0216 0313 0299 0144 0163 0079

Lu 0050 0025 0019 0059 0051 0034 0044 0045 0023 0027 0014

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1862

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 7: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 1 Continued

SW01 SW02 SW03 SW04 SW05 SW06 SW07 SW08 SW09 SW10 SW11 SW12

Lithology W Lz Lz Px Lz Lz Hz Lz Lz Lz Lz Lz

Modal compositions ()

Ol 587 517 349 637 608 766 607 728 626 649 567

Opx 210 352 151 230 271 178 221 177 223 208 252

Cpx 177 110 500 112 104 48 144 76 122 128 152

Sp 26 21 002 22 18 08 27 19 29 15 29

Fo 900 899 903 861 896 907 891 909 902 908 893

Major elements (wt )

SiO2 413 443 449 432 443 449 437 443 436 445 442 448

TiO2 030 008 006 008 009 011 002 012 002 007 006 009

Al2O3 229 329 307 319 270 293 121 346 179 308 222 358

TFe2O3 1423 818 894 760 951 842 813 888 813 805 795 854

MnO 018 012 010 010 013 011 010 013 009 011 009 012

MgO 348 377 372 260 388 399 441 386 429 400 405 380

CaO 505 375 236 926 264 238 109 302 162 262 278 331

Na2O 017 016 013 020 019 018 005 021 009 012 012 018

K2O 009 006 010 020 009 004 005 008 005 003 003 003

P2O5 003 002 003 003 002 003 003 005 003 004 003 002

LOI 105 188 248 969 098 040 095 048 102 078 142 079

Total 995 994 994 995 995 994 994 994 994 994 994 994

Trace elements (ppm)

Ni 1808 2412 2686 2363 2449 2343 2854 2294 2630 2391 2507 2266

Cu 471 298 240 324 27 332 124 120 141 269 206 412

Zn 934 537 612 574 751 600 537 632 595 562 524 524

Sr 111 758 592 2120 726 341 277 539 428 342 159 183

Y 346 282 214 254 280 291 0668 427 0882 231 167 324

Zr 169 248 230 604 659 995 339 117 404 209 400 225

Nb 0983 0083 0362 115 141 0972 0327 109 0231 0078 0445 0032

Ba 447 144 236 143 480 244 210 409 0985 144 156 0225

Hf 0453 0101 0084 0094 0166 0237 0055 0267 0075 0086 0101 0114

Ta 0061 0001 0005 0003 0050 0046 0017 0050 0020 0016 0016 mdash

Pb 0161 0117 0100 113 0083 0393 0636 0071 0173 0154 0248 0090

Th 0119 0015 0012 0020 0032 0045 0044 0107 0052 0039 0089 0006

U 0035 0012 0011 0019 0053 0021 0013 0033 0017 0013 0039 0004

S 550 50 80 60 550 80 60 550 550 80 90 270

Rare earth elements (ppm)

La 176 0052 0220 0262 128 129 103 249 0980 0082 0595 0053

Ce 404 0147 0380 0497 318 321 195 571 202 0168 108 0138

Pr 0554 0039 0053 0067 0421 0420 0203 0724 0219 0035 0122 0038

Nd 2688 0304 0299 0364 183 187 0731 325 0846 0254 0550 0329

Sm 0774 0173 0138 0158 0428 0436 0120 0726 0162 0134 0162 0194

Eu 0272 0074 0056 0073 0152 0160 0043 0247 0057 0061 0062 0084

Gd 0863 0303 0240 0270 0490 0526 0133 0818 0175 0261 0233 0352

Tb 0131 0061 0048 0054 0078 0084 0019 0122 0025 0051 0042 0076

Dy 0736 0450 0348 0399 0497 0527 0108 0755 0151 0376 0283 0549

Ho 0136 0104 0079 0091 0102 0112 0023 0151 0035 0087 0064 0123

Er 0335 0318 0251 0274 0304 0315 0068 0420 0101 0261 0194 0373

Tm 0041 0045 0038 0042 0042 0043 0010 0059 0014 0039 0028 0057

Yb 0246 0328 0268 0284 0291 0297 0070 0405 0097 0277 0194 0375

Lu 0034 0049 0043 0047 0044 0044 0011 0057 0016 0044 0030 0059

(continued)

CHU et al EASTERN NCC MANTLE THINING

1863

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 8: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 1 Continued

SW13 SW14 SW15 SW16 SW16-R SW17 SW18 SW19 SW20 SW21 SW22

Lithology Lz Lz Lz Lz Lz Lz Lz Lz W Lz

Modal compositions ()

Ol 650 615 615 663 578 716 603 731 635 613

Opx 221 214 237 227 262 162 161 158 30 257

Cpx 103 143 133 96 133 104 223 92 299 110

Sp 26 28 15 15 27 18 13 20 36 21

Fo 906 900 908 910 899 906 892 907 898

Major elements (wt )

SiO2 443 438 446 443 445 438 438 435 412 450

TiO2 007 010 006 003 009 004 010 005 048 007

Al2O3 270 340 239 205 353 194 285 213 375 289

TFe2O3 786 859 769 779 851 817 906 808 143 867

MnO 010 012 009 010 012 010 012 011 017 012

MgO 412 392 402 419 390 425 368 428 327 398

CaO 231 283 284 209 274 225 435 199 573 238

Na2O 013 014 011 011 017 009 022 009 022 013

K2O 003 005 003 005 008 003 005 003 007 005

P2O5 003 003 006 002 004 006 002 002 005 002

LOI 065 105 132 095 067 042 203 068 092 038

Total 994 993 994 994 994 994 994 994 995 995

Trace elements (ppm)

Ni 2311 2094 2571 2219 2279 1829 2254 2089 2292 1324 1929

Cu 272 298 243 262 270 878 274 132 188 638 256

Zn 556 601 520 476 483 603 487 582 479 952 679

Sr 164 650 793 122 123 643 170 383 315 105 226

Y 233 286 190 118 119 312 147 320 173 422 220

Zr 375 517 362 285 287 733 362 735 208 188 431

Nb 0318 0490 0134 0664 0670 101 0027 0545 0034 140 0237

Ba 145 130 108 223 228 469 0480 169 0348 697 0920

Hf 0116 0151 0086 0054 0049 0216 0079 0199 0083 0725 0131

Ta 0011 0026 0004 0041 0039 0061 0001 0031 0001 0091 0015

Pb 0410 0331 0108 0392 0383 0143 0106 0299 0053 0604 0057

Th 0024 0074 0071 0226 0222 0121 0008 0094 0005 0202 0042

U 0014 0024 0027 0081 0082 0038 0004 0033 0003 0070 0018

S 180 150 140 110 60 230 100 160 510 550

Rare earth elements (ppm)

La 0188 159 0397 101 0999 211 0144 0902 0032 294 0786

Ce 0466 352 0758 212 209 501 0533 186 0149 660 188

Pr 0075 0422 0094 0226 0217 0621 0093 0225 0038 0859 0239

Nd 0450 178 0457 0814 0801 267 0532 102 0266 420 109

Sm 0194 0425 0147 0159 0150 0595 0168 0314 0128 116 0287

Eu 0076 0148 0058 0054 0053 0209 0060 0122 0057 0409 0106

Gd 0303 0503 0228 0187 0182 0663 0222 0472 0235 134 0371

Tb 0058 0084 0041 0033 0031 0106 0040 0083 0046 0195 0067

Dy 0418 0540 0290 0224 0219 0663 0279 0586 0329 110 0439

Ho 0092 0115 0064 0052 0051 0138 0064 0127 0075 0197 0097

Er 0278 0338 0195 0161 0160 0407 0200 0385 0223 0481 0290

Tm 0041 0051 0029 0025 0024 0057 0030 0056 0033 0058 0043

Yb 0267 0329 0187 0173 0173 0387 0203 0392 0224 0343 0292

Lu 0043 0051 0030 0028 0028 0060 0033 0059 0036 0047 0045

Calculated using MINSQ (Hermann amp Berry 2002)Lz lherzolite Hz harzburgite D dunite W wehrlite Px pyroxenite R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1864

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 9: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

7500a system after digestion of samples using a mixture ofultra-pure HF and HNO3 in Teflon bombs (see AppendixB for analytical details) also at the Institute of Geologyand Geophysics Chinese Academy of Sciences Peridotitereference materials JP-1 and WPR-1 were measured tomonitor the accuracy of the analytical procedure and theresults are in good agreement with reference values(Appendix B Table B2) Precisions are generally betterthan 5 for most elements based on replicate analyses ofseveral samples (eg SW16 seeTable 1)The major element compositions of minerals were mea-

sured at the Institute of Geology and Geophysics ChineseAcademy of Sciences using a JEOL-JXA8100 electronmicroprobe operated in wavelength-dispersive (WDS)mode The operating conditions were as follows 15 kVaccelerating voltage counting time of 20 s and a 20 nAbeam current Natural minerals and synthetic oxides wereused as standards and a program based on the ZAF proce-dure was used for data correctionSulfur concentrations were determined at the National

Research Center for Geoanalysis Chinese Academy ofGeological Sciences using a high-frequency infraredabsorption spectrometer (HIR-944B Wuxi High-speedAnalyzer Co Ltd China) (Shi et al 2001) (seeAppendix B for method details) The quantification limitfor S of the method was about 50 ppm As Shi et al (2001)reported the analytical results for several Chinese rockreference materials using this method are comparablewith their nominal values Precisions are better than3 for samples with S concentrations of 4100 ppm andbetter than 10 for samples with S concentrations of5100 ppm

Clinopyroxene Sr^Nd^Hf isotope analysesStrontium Nd and Hf isotope compositions of cpx sepa-rates from the Penglai and Shanwang peridotite xenolithswere determined at the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences The mineralseparates were washed with ultra-pure (Milli-Q) waterand ground to 200^400 mesh using an agate mortarbefore isotopic analysis Analytical details for sampledigestion and column separation procedures are describedin Appendix B For comparison some cpx separates weretreated with the following steps and then reanalyzed forSr^Nd^Hf isotopes (1) the separate was leached with 6MHCl at 1008C overnight (2) the leached fraction was subse-quently rinsed with Milli-Q water (18 M) several times(3) the leached fraction was ground to 200^400 meshusing an agate mortarThe Rb^Sr and Sm^Nd isotopic analyses were con-

ducted using a Finnigan MAT 262 thermal ionizationmass spectrometer Measured 87Sr86Sr and 143Nd144Ndratios were corrected for mass-fractionation using86Sr88Srfrac14 01194 and 146Nd144Ndfrac14 07219 respectively

During the period of data collection the measured valuesfor the NBS-987 Sr standard and the JNdi-1 Nd standardwere 87Sr86Srfrac14 071024516 (2s nfrac14 8) and 143Nd144Ndfrac14 051211710 (2s nfrac14 8) respectively Lutetium and Hfwere measured using a ThermoElectron Neptune multi-collector ICP-MS system Hafnium isotopic ratios werenormalized to 179Hf177Hf frac14 07325 and 176Lu175Lu isotopicratios were normalized using Yb isotopic ratios Duringthe analytical campaign an Alfa Hf standard was mea-sured 10 times and the average value of 176Hf177Hf was0282179 4 (2s) The USGS reference material BCR-2was measured for Rb^Sr Sm^Nd and Lu^Hf isotopiccomposition to monitor the accuracy of the analytical pro-cedures with the following results 4654 ppm Rb3397 ppm Sr 87Sr86Sr frac14 070498613 (2s) 6676 ppmSm 2804 ppm Nd 143Nd144Nd frac14 051262313 (2s) and05175 ppm Lu 4912 ppm Hf 176Hf177Hf frac14 0282830 4(2s) These values are comparable with the reported refer-ence values (GeoREM httpgeoremmpch-mainzgwdgde) The procedural blanks were about 40 pg forRb 300 pg for Sr 20 pg for Sm 60 pg for Nd 20 pg for Luand 40 pg for Hf With the exception of sample PL19which has an Hf blank to sample ratio of about 05 ofwhich 176Hf177Hf is blank-corrected maximum blank tosample ratios are 002 for Sr (PL11) 009 for Nd(PL11) and 04 for Hf (PL11) requiring no correction ofthe measured isotopic ratios

Re^Os and PGE analysesRe^Os isotopic compositions and PGE abundances weredetermined at both the State Key Laboratory ofLithospheric Evolution Institute of Geology andGeophysics Chinese Academy of Sciences (IGGCAS) andthe Isotope Geochemistry Laboratory University ofMaryland (UMD) Some samples were analyzed in bothlaboratories for comparison As shown in Appendix B(Fig B1) the results from the two laboratories are in goodagreement for Os isotopic compositionsThe methods used at both laboratories are similar to

those described by Shirey amp Walker (1995) Pearson ampWoodland (2000) and Walker et al (2008) Rhenium OsIr Ru Pt Pd concentrations and Os isotopic compositionswere obtained from the same Carius tube sample digestion(for details of Re^Os and PGE chemistry see Appendix B)Osmium isotopic compositions were measured by nega-

tive thermal ionization using either a GV Isoprobe-Tmass spectrometer at IGGCAS (Chu et al 2007) or aVGSector 54 mass spectrometer at UMD (Walker et al 20022008) For these measurements purified Os was loadedonto platinum filaments and Ba(OH)2 was used as an ionemitter At IGGCAS all samples were run in static modeusing Faraday cups At UMD samples were run in staticmode on Faraday cups or in peak-jumping mode with asingle electron multiplier depending on the amount of OsThe measured Os isotopic ratios were corrected for mass

CHU et al EASTERN NCC MANTLE THINING

1865

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 10: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

fractionation using 192Os188Os frac14 30827 The in-run preci-sions for Os isotopic measurements were better than02 (2RSD) for all the samples During the period ofmeasurements of our samples the 187Os188Os ratio of theJohnson^Matthey standard of UMD was 011380 4 (2snfrac14 5) at IGGCAS and 0113792 (2s nfrac145) for Faradaycups and 011381 (2s nfrac14 5) for electron multiplierat UMDThe isotope dilution analyses of Re Ir Ru Pt and Pd

were conducted either at IGGCAS using a Thermo-Electron Neptune MC-ICP-MS system with an electronmultiplier in peak-jumping mode or using Faraday cups instatic mode according to the measured signal intensity orat UMD using a Nu-Plasma MC-ICP-MS system with atriple electron multiplier configuration in static modeMass fractionations (and gain effects of different multi-pliers for the UMD method) for Re Ir Ru Pt and Pdwere corrected using Re Ir Ru Pt and Pd standards thatwere interspersed with the samples In-run precisions for185Re187Re 191Ir193Ir 194Pt196Pt 105Pd106Pd and99Ru101Ru were typically 01^03 (2RSD)

RESULTSMajor element geochemistryThe intense serpentinization of the five xenoliths from theMengyin diamondiferous kimberlites is reflected in thehigh loss-on-ignition values (LOI from 105 to 140)(Table 1) Sample MY35 is characterized by high Al2O3Fe2O3 and CaO contents and relatively low MgO contentindicating that this sample may have originally been a pyr-oxenite The other four samples have relatively low Al2O3

and CaO contents Whole-rock compositions normalizedto 100 volatile-free are refractory with CaO rangingfrom 02 to 075wt and Al2O3 from 007 to 037wt similar to accepted values for the Archean cratonicmantle (eg Lee amp Rudnick 1999 Carlson et al 2005)The Mg-number of the whole-rocks ranges from 919 to952 with samples MY9 and MY34 having the highestwhole-rock Mg-number close to 95 Sulfur concentrationsin the Mengyin xenoliths are relatively high (from 140 to510 ppmTable 1)The 17 Penglai peridotites have low LOI values from03 to 05wt consistent with the generally lowdegrees of alteration of the xenoliths (Table 1 Fig 2)Some samples have negative LOI values indicating thatoxidation of FeO to Fe2O3 was more significant than lossof volatiles Whole-rocks are relatively fertile with Al2O3

ranging from 12 to 43wt and CaO ranging from 10to 33wt Most samples lie near the oceanic trend ofBoyd (1989) (Fig 3) which is considered typical of post-Archean peridotites Except for PL03 Al2O3 and CaOshow well-defined negative correlations with MgO content(Fig 4a and b) There are also negative correlationsbetween the Al2O3 content in whole-rocks and Fo and

Cr-number values of olivine and spinel (not shown)Sulfur concentrations in most Penglai samples are550 ppm (the quantification limit) except for samplesPL03 and PL16 which have S concentrations of 260 and60 ppm respectively (Table 1)The 22 Shanwang xenoliths also have relatively low LOI

(from 04 to 25wt Table 1) except for sample SW04for which LOI was 97wt Whole-rock xenoliths showvariable fertility as reflected by their Al2O3 contentswhich range from 12 to 38wt and CaO contentswhich range from 11 to 57wt Again most samples lienear the oceanic trend of Boyd (1989) (Fig 3) Al2O3 andCaO contents correlate negatively with the MgO contentexcept for samples SW01 SW04 and SW21 (Fig 4c andd) The latter samples have high CaO and relatively lowMgO contents reflecting their pyroxene-rich compositions(they are pyroxenites or wehrlites) Except for these sam-ples the negative linear correlations between the Al2O3

content of whole-rocks and Fo and Cr-number values ofolivine and spinel are also reasonably good (not shown)Sulfur contents are generally higher than those of Penglai(Table 1) with the wehrlite sample SW21having the highestS content of 510 ppm

Mineral chemistry and equilibrationtemperaturesMajor element compositions of minerals from Penglai andShanwang are provided in Table 2 Overall olivines haveFo-numbers ranging from 861 to 913 (Fig 3) The Fo-numbers of olivines from the Penglai peridotites rangefrom 881 to 913 (Table 2) with only two samples (PL03and PL16) having Fo-numbers lower than 89 Olivinesfrom the Shanwang xenoliths exhibit a similar range in Fo(890^910) with the exception of two samples havingvery low Fo (Fe-rich lherzolite SW05 with Fo of 861 andwehrlite SW21 with Fo of 818) Generally the mineralmodes and Fo compositions indicate that the Penglai andShanwang peridotites are relatively fertile lherzolitesThey are thus comparable with peridotite xenoliths fromother Cenozoic locales in eastern China and distinct fromthe very refractory xenoliths present in the PaleozoicMengyin and Fuxian kimberlites (Fan et al 2000Rudnick et al 2004 Zheng et al 2007) (Fig 3)The Cr-number of spinels from the Penglai and

Shanwang xenoliths range from 953 to 394 and from903 to 284 respectively (Table 2) never reaching thehigh Cr-number of spinels that are typical of cratonic peri-dotites and inclusions in diamonds (Cr-number460 egLee amp Rudnick 1999) as well as chromites from thePaleozoic kimberlites from eastern China (Zheng 1999)The equilibration temperatures of the mantle xenoliths

from Penglai and Shanwang can be constrained via appli-cation of various geothermometers (Wood amp Banno 1973Wells 1977 Bertrand amp Mercier 1985 Brey amp Kohler

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1866

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 11: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig 3 Modal olivine vs Mg-number of olivine in peridotites after Boyd (1989) Black diamond Fe-rich lherzolite SW05 from Shanwang smallopen diamonds literature data for Shanwang garnet-lherzolites from Zheng et al (2006)

Fig 4 Whole-rock Al2O3 and CaO vs MgO contents for Penglai (a b) and Shanwang (c d) peridotites Large open diamonds wehrlites andpyroxenites from Shanwang small open diamonds literature data for Shanwang peridotites from Zheng et al (2005) PM Primitive mantle(McDonough amp Sun1995) star average peridotitic xenoliths from Archean cratons (Griffin et al1999) gray field represents cratonic peridotitexenoliths from theTanzanian craton (Lee amp Rudnick 1999)

CHU et al EASTERN NCC MANTLE THINING

1867

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 12: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 2 Mineral compositions of peridotites from Penglai and Shanwang (nfrac14 3)

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL01 Ol 4112 000 001 000 988 040 014 4862 004 001 000 10022

Opx 5559 009 031 409 628 007 014 3253 052 010 001 9974

Cpx 5235 054 087 704 268 004 010 1425 1968 212 000 9968

Sp 001 009 1014 5643 1077 037 012 2012 000 000 001 9805

PL02 Ol 4102 000 002 000 880 036 012 4900 004 002 000 9938

Opx 5628 006 058 267 558 009 012 3337 069 006 000 9948

Cpx 5340 013 104 311 237 005 009 1654 2179 078 000 9929

Sp 006 066 3234 3343 1595 019 022 1574 000 003 000 9863

PL03 Ol 4071 001 001 001 1081 037 016 4745 007 000 000 9960

Opx 5446 008 040 527 684 007 018 3065 085 032 001 9911

Cpx 5209 047 078 705 336 004 010 1461 1833 228 000 9910

Sp 004 012 1050 5435 1256 037 013 1961 001 000 000 9770

PL06 Ol 4102 001 000 001 973 038 013 4841 002 000 001 9972

Opx 5485 015 028 430 605 009 015 3214 059 011 000 9871

Cpx 5208 059 071 673 260 002 010 1449 1976 192 001 9901

Sp 002 009 901 5737 1010 039 013 2066 000 000 000 9778

PL07 Ol 4079 002 000 006 991 036 014 4776 004 001 001 9908

Opx 5515 011 032 446 635 010 012 3208 061 009 000 9940

Cpx 5307 053 070 695 267 005 009 1507 1902 160 000 9976

Sp 009 009 906 5716 1020 040 012 2010 000 002 000 9724

PL08 Ol 4133 000 001 000 838 036 011 4953 004 001 000 9977

Opx 5629 003 050 334 530 007 016 3341 065 007 000 9980

Cpx 5306 016 116 454 212 004 006 1568 2126 127 000 9934

Sp 003 009 2164 4650 1047 029 015 1919 001 001 000 9838

PL13 Ol 4085 001 000 000 1008 036 015 4796 001 002 000 9945

Opx 5577 010 025 379 641 008 017 3257 045 008 000 9966

Cpx 5191 063 078 671 250 004 008 1400 2024 200 001 9889

Sp 002 007 918 5718 1082 036 013 1990 000 001 000 9767

PL14 Ol 4089 000 003 001 918 039 012 4871 004 000 000 9936

Opx 5531 007 045 420 584 007 013 3242 066 011 000 9926

Cpx 5261 045 098 611 236 003 008 1489 2005 183 000 9940

Sp 004 009 1398 5331 1004 032 011 1992 000 001 001 9784

PL16 Ol 4064 001 002 002 1131 029 014 4706 005 001 000 9954

Opx 5441 015 041 515 702 010 015 3076 088 016 000 9919

Cpx 5213 053 080 716 353 005 010 1477 1834 220 001 9962

Sp 007 019 1028 5433 1217 028 013 1969 000 001 000 9715

PL17 Ol 4102 001 002 002 930 036 013 4841 004 000 001 9929

Opx 5513 006 039 421 592 010 016 3207 065 011 000 9880

Cpx 5240 030 080 571 254 004 010 1525 2032 143 000 9889

Sp 003 004 1202 5463 1020 033 012 2001 000 002 001 9740

PL18 Ol 4105 000 002 001 891 040 013 4881 004 001 000 9937

Opx 5568 006 048 355 560 010 015 3273 066 006 001 9907

Cpx 5283 022 104 453 234 002 008 1585 2125 109 001 9928

Sp 001 008 2034 4717 1075 031 014 1894 000 000 000 9772

PL19 Ol 4098 001 001 000 920 041 014 4863 005 000 000 9944

Opx 5539 001 040 368 581 008 014 3233 062 008 001 9854

Cpx 5358 004 074 461 241 005 009 1587 2093 136 001 9968

Sp 006 001 1695 4855 1447 029 018 1716 001 000 001 9770

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1868

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 13: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

PL23 Ol 4123 000 000 001 858 040 013 4898 004 001 000 9940

Opx 5669 002 044 238 560 009 012 3358 053 001 000 9946

Cpx 5341 006 073 249 202 004 009 1666 2288 046 001 9885

Sp 002 655 1937 2039 4266 010 045 826 000 001 000 9780

SW02 Ol 4101 000 000 001 965 037 014 4852 002 001 000 9975

Opx 5543 008 029 374 620 011 016 3258 042 005 000 9906

Cpx 5262 049 070 597 230 004 010 1455 2095 179 000 9950

Sp 002 002 991 5741 1019 035 011 1982 000 001 000 9783

SW03 Ol 4084 000 001 000 977 035 013 4854 001 001 000 9965

Opx 5579 007 026 361 636 011 015 3266 044 004 000 9951

Cpx 5261 042 073 637 241 005 007 1426 2072 181 000 9944

Sp 009 003 902 5736 1005 035 011 2011 001 002 001 9716

SW04 Ol 4096 000 001 001 931 037 015 4855 001 003 000 9940

Opx 5507 009 043 441 603 008 016 3259 047 005 001 9940

Cpx 5233 039 083 593 235 004 008 1485 2067 164 000 9913

Sp 003 007 1136 5589 989 035 011 1992 000 001 000 9764

SW05 Ol 4033 000 001 002 1327 034 022 4606 005 001 001 10031

Opx 5509 010 038 393 809 008 021 3110 061 013 001 9972

Cpx 5266 040 081 421 293 004 007 1576 2104 125 001 9919

Sp 001 003 1198 5403 1191 038 012 1941 001 003 000 9791

SW06 Ol 4043 000 000 001 988 036 016 4800 005 000 000 9889

Opx 5464 011 034 416 605 013 016 3222 063 010 000 9855

Cpx 5216 060 091 659 250 006 009 1454 1974 199 001 9919

Sp 001 013 1170 5434 1072 038 011 2001 001 001 000 9740

SW07 Ol 4064 000 002 000 889 037 013 4877 003 002 001 9889

Opx 5612 006 053 329 574 004 013 3292 061 015 000 9961

Cpx 5351 016 140 485 246 004 009 1524 1912 219 000 9906

Sp 003 015 2510 4243 1180 024 019 1776 000 001 000 9772

SW08 Ol 4084 001 002 000 1049 038 015 4784 005 003 001 9983

Opx 5482 010 028 443 662 010 015 3180 066 013 001 9908

Cpx 5159 058 070 659 281 005 009 1454 1967 197 003 9863

Sp 002 004 907 5684 1085 038 012 2006 000 001 001 9740

SW09 Ol 4147 002 003 001 887 036 012 4944 008 004 000 10045

Opx 5564 006 055 330 558 008 014 3323 065 014 001 9936

Cpx 5372 013 139 504 241 004 008 1537 1909 216 000 9945

Sp 008 020 2356 4405 1134 027 018 1872 000 002 000 9842

SW10 Ol 4153 001 002 001 954 038 014 4908 003 001 001 10075

Opx 5598 003 027 341 614 012 013 3306 032 003 000 9949

Cpx 5236 043 074 584 234 003 008 1440 2089 176 000 9888

Sp 006 003 1024 5666 998 035 010 1983 001 000 000 9728

SW11 Ol 4092 000 002 000 894 039 013 4923 005 002 000 9969

Opx 5546 010 043 375 565 011 012 3291 063 009 001 9927

Cpx 5237 044 102 552 235 005 008 1518 2070 165 001 9937

Sp 002 013 1635 5122 1046 030 013 1999 000 002 001 9863

SW12 Ol 4084 001 001 000 1029 036 018 4810 002 001 000 9981

Opx 5581 008 025 389 667 006 013 3257 044 003 001 9993

Cpx 000 001 857 5791 1039 036 009 2018 001 001 001 9754

Sp 5219 048 063 621 245 004 009 1456 2108 169 001 9943

(continued)

CHU et al EASTERN NCC MANTLE THINING

1869

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 14: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

1990) a selection of which are shown in Table 3 Despitesystematic temperature differences between the thermo-meters the temperatures returned for each locality varyby no more than 3008C and the temperature ranges for agiven thermometer are indistinguishable between local-ities If the xenoliths equilibrated to a geotherm similar tothat proposed by Zheng et al (2006) for garnet-bearingShanwang peridotites the temperatures imply sampling

over a depth range of 20 km all within the spinel stabil-ity fieldAs mentioned above because of poor preservation of

primary minerals in the Mengyin peridotites their P^Tconditions cannot be constrained Minimum derivationdepths of 50^100 km however can be inferred from thepresence of garnet peridotites previously noted (Gao et al2002)

Table 2 Continued

Sample Mineral SiO2 TiO2 Cr2O3 Al2O3 FeO NiO MnO MgO CaO Na2O K2O Total

SW13 Ol 4125 000 001 001 910 034 013 4911 001 001 000 9997

Opx 5543 010 039 341 583 009 014 3335 039 006 001 9920

Cpx 5241 051 094 564 206 005 006 1479 2144 177 001 9970

Sp 001 009 1424 5361 1007 032 011 1972 000 001 001 9819

SW14 Ol 4097 000 003 003 971 037 013 4904 007 000 000 10035

Opx 5532 013 034 418 620 010 015 3255 062 012 001 9972

Cpx 5266 042 079 665 289 004 012 1554 1883 184 000 9977

Sp 003 015 1041 5621 1055 035 012 2040 001 002 000 9825

SW15 Ol 4133 001 001 001 901 038 012 5020 004 000 000 10111

Opx 5604 007 041 387 566 009 015 3369 063 010 001 10073

Cpx 5269 039 105 570 237 006 009 1541 2048 166 001 9991

Sp 004 012 1622 5220 1041 032 013 2031 000 001 001 9976

SW16 Ol 4146 000 001 000 891 037 013 5037 004 002 000 10133

Opx 5623 005 045 369 548 009 011 3366 062 013 001 10051

Cpx 5320 022 106 527 213 002 008 1552 2047 174 001 9970

Sp 004 007 1724 5173 1020 029 014 2034 000 001 000 10006

SW17 Ol 4112 000 000 000 991 039 014 4937 006 001 002 10102

Opx 5534 008 032 443 638 010 016 3274 060 013 000 10028

Cpx 5262 039 066 649 273 005 010 1502 1945 211 001 9962

Sp 002 006 999 5788 1060 039 011 2077 000 001 001 9984

SW18 Ol 4170 000 001 000 925 038 015 5013 001 000 000 10163

Opx 5655 007 040 323 591 007 015 3391 040 004 000 10074

Cpx 5353 034 101 508 235 004 006 1533 2139 155 000 10067

Sp 006 003 1749 5109 1086 030 013 1956 001 002 000 9954

SW19 Ol 4116 000 002 001 1050 037 013 4865 003 001 001 10087

Opx 5553 009 032 457 651 008 015 3247 062 010 000 10045

Cpx 5266 051 074 663 289 006 009 1513 1964 181 000 10016

Sp 004 012 974 5721 1059 040 013 2047 001 001 000 9872

SW20 Ol 4151 000 002 000 914 039 013 4969 002 001 000 10091

Opx 5649 009 034 348 586 011 013 3380 045 005 000 10082

Cpx 5302 047 091 585 218 003 007 1508 2085 172 001 10018

Sp 002 007 1389 5465 1002 033 013 2040 000 001 001 9952

SW22 Ol 4124 002 000 000 986 039 014 4855 005 000 000 10024

Opx 5581 007 034 420 610 013 016 3263 065 011 000 10021

Cpx 5291 043 090 630 263 003 006 1493 2001 182 000 10002

Sp 001 008 1211 5579 1040 038 012 2003 000 001 000 9894

n number of spot analyses for minerals total iron is expressed as FeO

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1870

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 15: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Rare earth and trace elementsAll Mengyin peridotites are strongly light REE (LREE)enriched (Fig 5a) and except for sample MY10 are char-acterized by relatively high total REE concentrations Ona primitive mantle-normalized diagram the Mengyinsamples show significant Sr depletion as well as variabledepletion inTh (Fig 5b)The REE patterns of the Penglai peridotites display little

variation (Fig 5c) The cpx-poor lherzolites (PL08 PL19)

and harzburgites (PL02 PL18 PL23) are relativelyLREE enriched and heavy REE (HREE) depletedSample PL19 is characterized by a spoon-shaped REE pat-tern and the two harzburgites PL02 and PL23 have thelowest HREE abundances of this suite LREE enrichmentof cpx-poor mantle xenoliths is common worldwide(McDonough amp Frey 1989) The cpx-rich lherzolites suchas PL01 PL06 PL07 PL16 and PL17 show more limitedLREE enrichment with the exception of sample PL03The LREE-depleted character of many of the Penglai cpxseparates (eight out of 19 samples Table 5 including mostof the lherzolites) suggests that the whole-rock REE pat-terns may have been compromised by addition of aLREE-enriched grain boundary phase (see Discussion)Compared with those from the Mengyin kimberlites thePenglai xenoliths have lower LaYb ratios Similar to theREE patterns the primitive mantle-normalized trace ele-ment patterns of the Penglai peridotites show relativelyuniform variations with slight enrichments in Sr anddepletions inTi (Fig 5d)The REE patterns of the Shanwang samples are more

variable (Fig 5e) The most refractory samples SW07 andSW09 are strongly enriched in LREE and depleted inHREE Lherzolites SW02 SW03 SW04 SW10 SW12SW13 SW18 and SW20 are LREE depleted albeit withsome showing La enrichment relative to Ce The remain-ing samples are LREE enriched The primitive mantle-normalized plots for the Shanwang samples are differentfrom those of the Penglai xenoliths showing pronouncedpositive Sr and negative Ti anomalies (Fig 5f) possiblyreflecting some form of Sr-rich (carbonate) melt interac-tion (Zheng et al 2005) The two wehrlites (SW01 andSW21) have the highest REE contents do not show theTiO2 depletions and show only modest Sr enrichments

Clinopyroxene Sr^Nd^Hf isotopic dataRb^Sr and Sm^Nd isotopic compositions of cpx from thePenglai and Shanwang xenoliths are provided in Tables 4and 5 respectively Most of the Penglai samples have87Rb86Sr ratios less than 006 and 87Sr86Sr ratios between07022 and 07042 except for unleached cpx PL15 whichhas a higher 87Rb86Sr of 011 (Fig 6a) The Shanwangsamples show even less variation in 87Rb86Sr and87Sr86Sr ratios (Fig 6a) Generally the leached cpx areslightly less radiogenic in 87Sr86Sr compared with theirpaired unleached cpx (Table 4 Fig 6a) Although the dif-ferences in Sr concentration between leached andunleached cpx are insignificant the Rb concentrations inleached cpx are much lower than their paired unleachedcpx (Table 4 Fig 6a) Similar to other worldwide xenolithsuites no Rb^Sr isochron can be generated from the dataPenglai and Shanwang samples show modest variations in147Sm144Nd and 143Nd144Nd ratios (Fig 6b) The leachedcpx have 147Sm144Nd and 143Nd144Nd ratios nearly identi-cal to their paired unleached cpx except for samples

Table 3 Temperature calculations for Penglai and

Shanwang peridotites using various geothermometers

Wood amp Wells Bertrand amp Brey amp

Banno (1973) (1977) Mercier (1985) Kohler (1990)

PL01 1055 958 1071 1005

PL02 1069 963 1019 965

PL03 1110 1038 1178 1094

PL06 1065 967 1075 1012

PL07 1127 1047 1176 1113

PL08 1059 947 1008 956

PL13 1014 910 999 935

PL14 1072 973 1069 1012

PL16 1113 1044 1188 1101

PL17 1080 984 1074 1014

PL18 1067 962 1028 973

PL19 1075 976 1052 996

PL23 1009 893 910 855

SW02 1002 894 963 903

SW03 1009 904 981 918

SW04 1037 933 1010 952

SW05 1036 961 1036 970

SW06 1064 966 1074 1013

SW07 1111 1016 1132 1075

SW08 1056 965 1073 1004

SW09 1119 1022 1143 1088

SW10 998 888 954 892

SW11 1050 943 1020 963

SW12 992 888 952 888

SW13 974 856 899 839

SW14 1134 1053 1181 1115

SW15 1072 967 1057 1001

SW16 1077 970 1055 1005

SW17 1084 993 1110 1046

SW18 1028 918 984 926

SW19 1093 1007 1123 1056

SW20 1041 933 1010 956

SW22 1078 982 1087 1024

Pressure of 15 GPa assumed throughout

CHU et al EASTERN NCC MANTLE THINING

1871

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 16: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig 5 Chondrite-normalized rare earth element patterns (a c e) and primitive mantle-normalized trace element patterns (b d f) for thexenoliths from Mengyin Penglai and Shanwang Chondrite normalizing values are from Masuda et al (1973) divided by 12 and those of prim-itive mantle for trace elements are from McDonough amp Sun (1995) R replicate analysis

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1872

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 17: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 4 Cpx Rb^Sr isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Rb (ppm) Sr (ppm) 87Rb86Sr 87Sr86Sr 2s ISr

PL01 5 0190 7170 000767 0702319 0000014 0702318

PL02 5 0408 1392 000847 0702696 0000014 0702695

PL03 5 0116 1555 000215 0703476 0000020 0703476

PL03-leached 0109 1646 000191 0703373 0000007 0703373

PL06 5 0338 7545 00130 0702646 0000013 0702645

PL07 5 0207 5444 00110 0702570 0000014 0702569

PL11 5 0266 1318 00584 0704250 0000046 0704245

PL12 5 0168 8995 000541 0702552 0000010 0702551

PL13 5 0359 3079 000338 0703293 0000014 0703292

PL14 5 0492 7513 00190 0702897 0000014 0702896

PL15 5 219 5909 0107 0702831 0000014 0702824

PL15-leached 0100 6058 000479 0702795 0000027 0702795

PL16 5 0271 1602 000490 0703339 0000012 0703339

PL17 5 0257 3496 00212 0703335 0000013 0703334

PL17-leached 00983 3288 000864 0702914 0000010 0702914

PL18 5 0346 2021 00495 0704127 0000014 0704123

PL19 5 00766 4113 000539 0703623 0000013 0703622

PL19-leached 3753 0703496 0000014

PL19-leached-R 00173 3740 000133 0703470 0000011 0703470

PL23 5 0308 3426 00260 0703646 0000013 0703644

PL23-leached 0094 3297 00083 0703590 0000016 0703590

SW01 16 222 1177 00546 0703220 0000013 0703207

SW01-leached 00806 1003 000232 0703053 0000013 0703053

SW02 16 0350 2021 00500 0702676 0000013 0702664

SW03-leached 16 00230 1051 000632 0702895 0000016 0702893

SW04 16 0447 2887 00463 0703135 0000018 0703124

SW04-leached 00074 2002 00011 0702450 0000014 0702450

SW05 16 0569 1650 00100 0703183 0000015 0703181

SW05-leached 00525 1617 0000938 0703031 0000011 0703031

SW07 16 0860 2774 000896 0703131 0000013 0703129

SW08 16 149 1468 00294 0703391 0000015 0703384

SW09 16 0845 2829 000863 0703145 0000013 0703143

SW10 16 0372 2318 00464 0702631 0000013 0702620

SW11 16 0264 9052 000845 0703176 0000013 0703174

SW11-leached 00148 8226 0000519 0703158 0000017 0703158

SW12 16 0231 1444 00463 0702590 0000036 0702579

SW13 16 0481 2793 00498 0702665 0000012 0702654

SW14 16 0469 1276 00106 0703132 0000014 0703129

SW15 16 0307 1046 000849 0703259 0000011 0703257

SW16 16 0626 1765 00103 0703525 0000014 0703522

SW17 16 0260 2243 000335 0703133 0000012 0703132

SW18 16 0048 6099 000228 0702268 0000013 0702268

SW19 16 0261 1099 000688 0702905 0000013 0702904

SW20 16 0279 2924 00276 0702499 0000013 0702493

SW21 16 0622 1260 00143 0703100 0000013 0703097

SW22 16 1675 0703064 0000012

SW22-leached 00878 1629 000156 0703041 0000013 0703041

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh lsquoRrsquoreplicate analysis

CHU et al EASTERN NCC MANTLE THINING

1873

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 18: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 5 Cpx Sm^Nd isotopic compositions of the Penglai and Shanwang peridotites

Sample t (Ma) Sm (ppm) Nd (ppm) 147Sm144Nd 143Nd144Nd 2s eNd(t) eNd(0) TDM (Ma)

PL01 5 1349 3391 02406 0513215 0000014 114 113 364

PL02 5 1082 4399 01487 0513003 0000012 73 71 347

PL03 5 3291 1320 01508 0512983 0000010 69 67 408

PL03-leached 4161 1696 01483 0512977 0000008 67 66 406

PL06 5 1463 3725 02375 0513080 0000012 87 86 457

PL07 5 1340 3254 02490 0513168 0000013 105 103 76

PL11 5 01487 05354 01679 0512971 0000040 66 65 601

PL12 5 1681 4176 02433 0513176 0000015 106 105 128

PL13 5 2614 1244 01270 0512908 0000012 54 53 428

PL14 5 1313 3528 02251 0513285 0000010 127 126 1794

PL15 5 1102 2781 02396 0513190 0000014 109 108 232

PL15-leached 1246 3202 02355 0513203 0000007 111 110 362

PL16 5 2214 7993 01674 0513022 0000013 76 75 425

PL17 5 07616 1403 03283 0513813 0000013 231 229 881

PL17-leached 08447 1491 03427 0513867 0000009 241 240 847

PL18 5 07417 1424 03150 0513406 0000013 151 150 384

PL19 5 02420 1197 01222 0513054 0000015 82 81 162

PL19-leached 02335 1125 01255 0513143 0000010 100 98 14

PL19-leached-R 02303 1138 01224 0513126 0000011 96 95 42

PL23 5 02675 09223 01754 0512795 0000035 32 31 1412

PL23-leached 02559 08949 01732 0512894 0000008 51 50 967

SW01 16 2364 8035 01778 0512944 0000013 64 60 878

SW01-leached 2882 9875 01765 0512964 0000013 68 64 766

SW02 16 09915 1678 03571 0513411 0000011 155 151 277

SW03 16 1031 1673 03725 0513018 0000012 78 74 128

SW03-leached 1174 1927 03684 0513076 0000007 89 85 74

SW04 16 1029 1834 03391 0513479 0000010 168 164 400

SW04-leached 1155 2070 03379 0513473 0000007 167 163 396

SW05 16 2248 8976 01514 0513078 0000012 90 86 179

SW07 16 2475 1285 01164 0512940 0000013 63 59 331

SW08 16 2360 9530 01497 0512981 0000013 71 67 405

SW09 16 2415 1213 01204 0512940 0000013 63 59 346

SW10 16 1051 1796 03537 0513457 0000012 164 160 334

SW11 16 1163 3262 02156 0513123 0000009 99 95 2301

SW11-leached 1339 3830 02114 0513105 0000007 95 91 3079

SW12 16 1057 1742 03669 0513411 0000014 155 151 259

SW13 16 1204 2388 03048 0513266 0000012 127 123 193

SW14 16 1551 5989 01566 0513095 0000011 93 89 150

SW15 16 1188 3809 01886 0513090 0000013 92 88 370

SW16 16 09280 4575 01226 0513078 0000012 90 86 123

SW17 16 2206 9757 01367 0512982 0000012 71 67 336

SW18 16 1240 3605 02079 0513291 0000010 131 127 3742

SW19 16 1492 4852 01859 0513202 0000012 114 110 281

SW20 16 1197 2370 03053 0513267 0000012 127 123 194

SW21 16 2733 9033 01829 0512986 0000013 72 68 815

SW22-leached 16 2342 9970 01421 0513039 0000005 82 78 240

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400mesh R replicate analysis eNd values were calculated using (147Sm144Nd)CHUR(0)frac14 01967 and(143Nd144Nd)CHUR(0)frac14 0512638 TDM values were calculated using (147Sm144Nd)DM(0)frac14 02137 and(143Nd144Nd)DM(0)frac14 0513151

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1874

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 19: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig 6 Rb^Sr (a) Sm^Nd (b) and Lu^Hf (c) isotopic compositions of the cpx from the Penglai and Shanwang peridotites Open circles lea-ched Penglai cpx open diamonds leached Shanwang cpx In (c) the Lu^Hf isochron with an age of 125922 Ma (using leached PL19 cpxdata) is for Penglai samples only (although Shanwang data are also plotted in the figure) If the Penglai and Shanwang xenoliths are combinedan isochron age of 126235 Ma (using leached PL19 cpx data) is obtained (not shown) The inset shows the data for samples with 176Lu177Hf501

CHU et al EASTERN NCC MANTLE THINING

1875

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 20: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

PL19 and PL23 which have slightly more radiogenic Nd inthe leached cpx (Table 5 Fig 6b) Except for samplesPL17 and SW03 all samples plot near an errorchron withan age of 300 Ma (Fig 6b)Hafnium isotopic compositions of the cpx are provided

inTable 6 and shown in Fig 6c The Penglai samples showlarger ranges of 176Lu177Hf and 176Hf177Hf ratios than theShanwang xenoliths Generally the leached cpx have Hfisotopic compositions that are indistinguishable from thepaired unleached cpx although there are some significantdifferences in Lu and Hf concentrations (Table 6) Theonly exception is sample PL19 which is characterized byextremely radiogenic Hf the leached cpx has slightlymore radiogenic Hf (Fig 6c) The 176Hf177Hf of two analy-ses of leached PL19 cpx are reproducible within uncertain-ties Penglai cpx samples define a Lu^Hf isochron with anage of 125922 Ma (using the data for the leached PL19cpx) (Fig 6c) If the Penglai and Shanwang xenoliths arecombined an isochron age of 126235 Ma is obtained(also using data for leached cpx from sample PL19)All the cpx samples are characterized by low initial

87Sr86Sr ratios of 07022^07042 and high initial eNd andeHf values of thorn54 to thorn231 and thorn160 to thorn553 when cor-rected for the eruption ages of the host basalts (Fig 7) Acrude negative correlation between 87Sr86Sr and eNd issimilar to cpx from other Cenozoic mantle xenoliths ineastern China (Fig 7a) (Fan et al 2000 Rudnick et al2004 Wu et al 2006 and references therein) These Sr^Nd isotopic compositions differ greatly from those for thePaleozoic Mengyin and Fuxian kimberlites and themantle xenoliths they host (Chi amp Lu 1996 Zheng 1999Wu et al 2006 Zhang amp Yang 2007 Zhang et al 2008)which are characterized by much more enriched Sr^Ndisotope signatures (Fig 7a) In a diagram of eNd vs eHfmost xenoliths straddle a line of eHffrac14 15eNd which is sim-ilar to the Nd and Hf isotopic ratios seen in oceanic basalts(Nowell et al 1998 Vervoort et al 1999) However somesamples such as PL19 and PL11 fall above the line atmuch higher eHf values (Fig 7b) The apparent decouplingof the Nd^Hf systematics in these samples suggests thathigh LuHf was preserved for a much longer period oftime than high SmNd Leached cpx sample PL19 whichhas an extremely high 176Lu177Hf ratio of about 076yields a Hf model age of about 13 Ga

Re^Os isotopic geochemistryRe^Os isotopic data for the Mengyin Penglai andShanwang xenoliths are given in Table 7 All Mengyinsamples have Re concentrations less than 01ppb with alarge range in Os concentrations from 01 to 78 ppb(Fig 8a) Three analyses of sample MY9 yielded Os con-centrations ranging between 20 and 78 ppb indicating asignificant lsquonugget effectrsquo whereby Os is evidently concen-trated in trace phases that are not well homogenizedduring the preparation of sample powders Peridotites

MY9 MY10 and MY34 have 187Os188Os that are identicalwithin uncertainties 01104^01109 corresponding to TRD

ages of 25^26 Ga (Table 7) These model ages are similarto ages reported for peridotites from the Fuxian andTieling kimberlites (Gao et al 2002 Wu et al 2006Zhang et al 2008) and chromites from Mengyin (Wuet al 2006) In contrast to the other Mengyin peridotitesMY33 has a higher 187Os188Os ratio of 01166 correspond-ing to aTRD age of 16 Ga (Table 7) In addition pyroxe-nite sample MY35 is characterized by relatively low Os(097 ppb) relatively high 187Re188Os (0288) and a high187Os188Os ratio of 01319 Relatively high ratios like thisare common in pyroxenites from the lithospheric mantle(eg Becker et al 2004)The Penglai peridotites are strongly depleted in Re

with concentrations ranging from5001ppb to 004 ppbOs concentrations are mostly 51ppb (Fig 8a) The187Re188Os ratios range from 001 to 023 As is commonin other peridotite xenoliths from the continental litho-spheric mantle the 187Re188Os ratios do not correlate wellwith their 187Os188Os ratios (Fig 8b) Generally thePenglai samples have 187Os188Os ratios ranging from01178 to 01289 which is lower than the primitive uppermantle (PUM) estimate of 01296 (Meisel et al 2001)Among these relatively refractory samples PL02 PL08PL12 and PL23 (52wt Al2O3) have similar 187Os188Osratios around 012 with TRD ages of 1 Ga (Table 7)Sample PL18 however which has a slightly higher Al2O3

of 227wt has the lowest 187Os188Os of 01170 withTRD and TMA ages of 14 and 25 Ga respectively(Table 7)Shanwang xenoliths have higher average Re and Os

concentrations than the Penglai xenoliths with Re of001^051ppb and Os of 03^ 87 ppb (Fig 8a) The excep-tion is Fe-rich lherzolite SW05 which has unusually lowRe and Os concentrations of 0005 and 004 ppb respec-tively with high 187Re188Os and 187Os188Os of 0630 and01352 respectively Lherzolites SW12 and wehrlites SW01and SW21 have higher 187Re188Os ratios of 0589^1636than PUM and have correspondingly chondritic to radio-genic 187Os188Os of 01265^01335 (Fig 8b)The remainingsamples have 187Os188Os ratios of 01196^01274 Of theserelatively refractory samples SW07 SW09 SW11 SW15SW16 and SW20 have low 187Os188Os ratios rangingfrom 0120 to 0122 withTRD ages ranging from 08 to 11Ga (Table 7) Sample SW03 although not refractory(32wt Al2O3) also has a

187Os188Os of 012 (Table 7)

Platinum-group elementsThe PGE abundances (especially Ir Ru) in replicate anal-yses of some samples vary considerably beyond analyticaluncertainties (Table 8) indicating that the lsquonugget effectrsquois significant for the Mengyin Penglai and Shanwangxenoliths Despite this the chondrite-normalized PGE

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1876

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 21: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 6 Cpx Lu^Hf isotopic compositions of the Penglai and Shanwang peridotites

Sample t Lu Hf 176Lu176Hf 176Hf177Hf 2s eHf(0) eHf(t) TDM

(Ma) (ppm) (ppm) (Ma)

PL01 5 0215 0984 00311 0283283 0000009 181 181 244

PL02 5 00940 0477 00280 0283332 0000017 198 198 423

PL03 5 0167 0958 00248 0283432 0000007 234 234 722

PL03-leached 5 0207 108 00274 0283426 0000005 231 232 865

PL06 5 0208 103 00288 0283277 0000007 179 179 153

PL07 5 0217 0922 00334 0283265 0000007 174 174 159

PL11 5 0111 00915 0172 0286727 0000062 140 139 1377

PL12 5 0105 0581 00257 0283250 0000013 169 169 1

PL13 5 0248 111 00318 0283225 0000006 160 160 199

PL14 5 0194 0859 00321 0283456 0000009 242 242 1779

PL15 5 0177 0699 00360 0283365 0000012 210 210 2614

PL15-leached 5 0213 0753 00402 0283396 0000005 221 220 4192

PL16 5 0165 114 00205 0283298 0000008 186 186 143

PL17 5 0175 0420 00593 0284162 0000017 491 491 2285

PL17-R 5 0171 0427 00569 0284205 0000016 507 506 2694

PL17-leached 5 0209 0431 00690 0284259 0000007 526 525 1740

PL18 5 0156 0582 00381 0283783 0000012 358 358 mdash

PL19 5 0130 00267 0693 0298410 0000177 553 551 1227

PL19-leached 5 0121 00232 0747 0301364 0000048 654 652 1345

PL19-leached-R 5 0125 00232 0768 0301392 0000058 655 652 1308

PL23 5 00653 0133 00696 0284091 0000075 467 465 1424

PL23-leached 5 00612 0129 00675 0284299 0000019 540 539 1895

SW01 16 00974 149 000928 0282981 0000010 74 77 492

SW02 16 0202 0605 00474 0283297 0000014 186 184 275

SW03 16 0221 0618 00509 0283097 0000013 115 113 660

SW03-leached 16 0271 0659 00584 0283098 0000004 115 113 409

SW04 16 0224 0591 00538 0283544 0000016 273 271 1011

SW04-leached 16 0252 0621 00577 0283575 0000005 284 281 895

SW05 16 0222 0887 00356 0283324 0000011 195 195 1457

SW05-leached 16 0239 0944 00360 0283318 0000006 193 193 1548

SW07 16 0121 0821 00209 0283327 0000012 196 198 236

SW08 16 0215 0989 00309 0283104 0000010 117 118 1028

SW09 16 0114 0838 00193 0283327 0000011 196 198 218

SW10 16 0238 0611 00553 0283371 0000019 212 209 380

SW11 16 0168 0704 00340 0283410 0000011 225 225 1982

SW11-leached 16 0194 0774 00356 0283405 0000005 224 224 3107

SW12 16 0244 0584 00594 0283302 0000013 188 185 133

SW13 16 0210 0691 00432 0283327 0000011 196 195 852

SW14 16 0214 0633 00480 0283448 0000011 239 238 1097

SW15 16 0162 0692 00334 0283446 0000014 238 238 2121

SW16 16 0169 0256 00941 0284878 0000026 745 738 1543

SW17 16 0221 0735 00428 0283249 0000012 169 168 12

SW18 16 0188 0801 00335 0284044 0000013 450 450 9369

SW19 16 0216 0974 00315 0283454 0000012 241 241 1614

SW20 16 0213 0709 00427 0283320 0000015 194 193 870

SW21 16 0102 171 00085 0282999 0000010 80 83 447

SW22 16 0194 0867 00318 0283641 0000010 307 308 3255

SW22-leached 16 0215 0884 00346 0283666 0000005 316 316 6178

lsquoleachedrsquo indicates that the minerals were leached with 6M HCl at 1008C overnight before crushing to 200ndash400 mesh Rreplicate analysis eHf values were calculated using (176Lu177Hf)CHUR(0)frac14 00332 and (176Hf177Hf)CHUR(0)frac14 0282772TDM values were calculated using (176Lu177Hf)DM(0)frac14 00384 and (176Hf177Hf)DM(0)frac14 028325 Hf isotopic ratios of repli-cate analyses of sample PL19 are blank-corrected (using blank values Hf 40 pg 176Hf177Hf ratio of 0283)

CHU et al EASTERN NCC MANTLE THINING

1877

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 22: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig 7 Sr^Nd (a) and Nd^Hf (b) isotope correlation diagrams for cpx from the Penglai and Shanwang peridotiteslsquotrsquo represents the eruptionage of the host basalts Open circles leached Penglai cpx open diamonds leached Shanwang cpx In (a) data sources for Paleozoic kimberlitesand their mantle xenoliths are Chi amp Lu (1996) Zheng (1999)Wu et al (2006) Zhang amp Yang (2007) and Zhang et al (2008) data sources forCenozoic mantle xenoliths from the eastern NCC are Fan et al (2000) Rudnick et al (2004)Wu et al (2006) and references therein

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1878

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 23: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 7 Re^Os isotopic compositions of the mantle xenoliths

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

MY9 470 037 0068 198 017 011052 007 118 010921 260 405

MY9-R 470 037 0089 780 0055 011039 005 112 010996 249 281

MY9 470 037 0017 426 0020 011056 005 109 011040 243 253

MY10 470 006 0005 0090 03 011091 031 121 010886 265 644

MY33 470 031 0008 119 003 011662 005 60 011635 157 167

MY34 470 092 0050 330 0073 011043 005 113 010986 251 295

MY35 470 100 0058 0969 029 013195 010 47 012969 040 268

PL01 5 358 898 108 1005 0008 0280 01 012544 12 12 012543 023 036

PL02 5 120 908 394 965 0011 0654 0081 011974 026 57 011974 108 134

PL02 5 0005 0835 003 012103 012 47 012103 089 096

PL03 5 401 887 115 1094 0031 0710 021 012677 013 02 012675 004 007

PL03-R 5 0031 0651 023 012588 006 09 012586 017 039

PL03 5 0523 115 219 012618 021 08 012599 015 003

PL06 5 433 899 953 1012 0019 0400 023 012608 037 07 012606 014 032

PL07 5 399 896 962 1113 0181 0829 105 012697 005 01 012688 002 000

PL07 5 0032 0751 020 012722 011 02 012720 003 007

PL08 5 170 913 238 956 0019 0496 019 012032 027 52 012030 099 183

PL08 5 0001 0426 001 012060 008 50 012060 095 098

PL10 5 324 0020 122 0080 012272 006 34 012271 064 079

PL11 5 279 0015 0747 0098 012519 010 14 012518 027 036

PL11 5 0020 101 0097 012479 008 17 012478 033 043

PL12 5 186 0028 120 011 012072 003 49 012071 093 129

PL13 5 391 895 972 935 0009 0406 01 012885 013 15 012884 028 038

PL14 5 347 904 150 1012 0011 0411 013 012754 013 04 012753 008 012

PL14 5 0008 0470 008 012529 004 13 012529 026 032

PL15 5 247 0014 0680 0098 012136 015 44 012135 084 110

PL16 5 386 881 113 1101 0009 0593 007 012599 006 08 012598 015 018

PL16-R 5 0065 0729 043 012596 018 08 012592 016 227

PL16 5 0011 115 0046 012588 010 09 012587 017 019

PL17 5 369 903 129 1014 0012 0541 010 012592 010 08 012591 016 022

PL17 5 0015 0688 011 012546 010 12 012545 023 031

PL18 5 227 907 224 973 0072 0476 073 011765 014 74 011759 139 172

PL18-R 5 0020 0622 016 011672 009 81 011670 152 248

PL18 5 0410 102 193 011741 005 77 011725 144 038

PL19 5 244 904 190 996 0038 0810 023 012711 023 01 012709 001 004

PL19 5 0030 104 014 012533 006 13 012532 025 038

PL19-R 5 0036 0833 021 012547 022 12 012546 023 047

PL23 5 188 911 389 855 0012 121 0049 012029 007 53 012028 099 113

PL23 5 0011 111 0046 012077 004 49 012077 092 104

SW01 16 233 0072 0282 12 013351 036 50 013319 093 047

SW02 16 337 900 104 903 0161 299 0259 012611 004 07 012605 014 037

SW03 16 317 899 955 918 0100 321 0150 012101 002 47 012097 089 141

SW04 16 355 903 120 952 0032 222 0068 012278 007 33 012276 063 075

SW05 16 274 861 130 970 0005 0036 06 013520 130 64 013503 121 212

SW06 16 296 896 126 1013 0077 202 018 012416 005 22 012411 043 078

SW07 16 123 907 284 1075 0420 139 144 012219 003 40 012181 077 028

(continued)

CHU et al EASTERN NCC MANTLE THINING

1879

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 24: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

patterns (and hence PGE element ratios) are generallyreproducibleFor the Mengyin samples the chondrite-normalized

PGE patterns show relatively flat IPGE (Os Ir Ru) pat-terns with corresponding strong PPGE (Pt Pd) depletions(Fig 9a) Thus (OsIr)N and (RuIr)N of Mengyin perido-tites straddle unity (Fig 9d) but (PtIr)N and (PdIr)Nvalues are uniformly less than 1 (Fig 9e and 9f) Thesecharacteristics are frequently observed in ancient cratonicperidotite xenoliths from elsewhere (Rehkalaquo mper et al1997 Pearson et al 2004) and have previously beenascribed to large degrees of melt depletion (Pearson et al2004)The Penglai peridotites show Os depletion relative to Ir

and are also depleted in Pt and Pd (Fig 9b) Their (OsIr)N range from 013 to 063 much lower than those fromthe Mengyin xenoliths (Fig 9d) Similar dramatic fractio-nation of Os from Ir has been noted in peridotitic xenolithsfrom Vitim Siberia (Pearson et al 2004) and NorthQueensland Australia (Handler et al 1999) and has pre-viously been attributed to Os loss due to syn- or post-eruption sulfide breakdown and alteration (Handler et al1999 Pearson et al 2004) Moreover the (PtIr)N and(PdIr)N values of the Penglai xenoliths are mostly lessthan unity (Fig 9e and f) Harzburgite PL18 has the

lowest (PdIr)N value consistent with its unradiogenic Osisotopic ratioThe Shanwang xenoliths show large variations in PGE

concentrations (Table 8) In contrast to peridotites fromMengyin and Penglai this suite of xenoliths does notshow significant fractionation between IPGE and PPGE(Fig 9c) Although the Shanwang xenoliths have almostthe same (OsIr)N values as those from Mengyin their(RuIr)N (PtIr)N and (PdIr)N values are much higher(Fig 9e and f) and some samples have PGEthornRe patternssimilar to estimates of PUM (eg Becker et al 2006)

DISCUSS IONLimitations to dating lithospheric mantleusing the Re^Os isotopic systemIt has been shown that Re^Os isotopic systematics can beused to date melt extraction from the mantle and thus theage of formation of the lithospheric mantle (Shirey ampWalker 1998) The principle is that the daughter elementOs is strongly compatible in mantle residues whereas theparent element Re is moderately incompatible duringmantle melting (Walker et al 1989) Removal of Re accom-panying melt depletion leads to retardation or cessation inthe growth of 187Os Because of the lack of the parent

Table 7 Continued

Sample T (Ma) Al2O3

(wt )

Fo

(Ol)

Cr-no

(Sp)

T

(8C)

Re

(ppb)

Os

(ppb)

187Re188Os 187Os188Os 2s gOs Osi TRD

(Ga)

TMA

(Ga)

SW07 16 0012 138 0041 012230 008 36 012229 070 078

SW08 16 350 890 967 1004 0030 0419 034 012739 016 03 012729 004 041

SW09 16 182 909 264 1088 0050 145 016 012171 021 41 012167 079 128

SW09 16 0028 233 0058 012209 010 38 012208 073 085

SW10 16 312 902 108 892 0224 327 0331 012395 003 24 012386 047 251

SW11 16 227 908 176 963 0159 483 0159 012131 002 44 012127 085 139

SW12 16 363 893 903 888 0508 415 0589 012979 004 22 012964 039 089

SW13 16 273 906 151 839 0312 438 0343 012292 007 32 012283 062 402

SW14 16 346 900 110 1115 0262 388 0326 012515 004 14 012506 029 144

SW15 16 244 909 172 1001 0248 451 0265 012138 008 44 012131 084 241

SW16 16 208 910 183 1005 0263 869 0146 011996 001 55 011992 105 163

SW17 16 357 899 104 1046 0113 353 0154 012325 005 29 012321 056 090

SW18 16 196 906 187 926 0415 483 0414 012501 001 16 012490 031 1106

SW18 16 0362 438 0398 012527 011 14 012516 027 2071

SW19 16 293 892 103 1056 0083 250 016 012500 003 15 012496 030 049

SW20 16 216 907 145 956 0200 412 0234 011965 002 58 011959 110 257

SW21 16 380 818 210 0332 0978 164 012650 006 07 012606 014 002

SW22 16 292 898 127 1024 0025 0564 021 012255 009 35 012249 067 139

The parameters used in calculation are Refrac14 1666 10ndash11year (187Re188Os)Chond(0) frac14 040186 (187Os188Os)Chond(0)frac14 01270 (Shirey amp Walker 1998) R replicate analysis Al2O3 (wt ) is normalized to 100 volatile-free Re Osconcentration and Os isotopic ratios are blank-correctedMeasured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1880

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 25: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

isotope the residue is relatively immune to subsequent dif-fusive resetting at high temperatures in mantle peridotitesThe most robust method to date mantle melt depletion

using the Re^Os system would be via the generation of

whole-rock isochrons This method however is generallynot viable for peridotites because of the presumptionof lack of isotopic homogeneity in a large mantle domainat the time of melting and the possible late-stage mobility

Fig 8 Re^Os (a) and 187Re188Os^187Os188Os (b) variation diagrams for the Mengyin Penglai and Shanwang xenoliths (a) PUM (opensquare) values from Becker et al (2006) (b) the larger panel shows a close-up of the data with 187Re188Os up to 05 and the inset shows thefull dataset PUM (open square) values from Meisel et al (2001) open diamonds wehrlites SW01 and SW21 and Fe-rich lherzolite SW05 fromShanwang that show evidence of metasomatism

CHU et al EASTERN NCC MANTLE THINING

1881

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 26: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Table 8 PGE elemental compositions of the mantle xenoliths

Samples Os Ir Ru Pt Pd Re (OsIr)N (RuIr)N (PtIr)N (PdIr)N

(ppb) (ppb) (ppb) (ppb) (ppb) (ppb)

MY9 780 552 511 076 043 0089 131 059 006 006

MY9 426 318 527 044 033 0017 124 106 006 008

MY10 009 007 012 003 000 0005 126 119 019 006

MY33 119 148 163 015 009 0008 075 070 005 005

MY34 330 348 527 220 025 0050 088 097 029 006

MY35 097 096 151 034 060 0058 093 100 016 052

PL01 028 199 323 278 094 0008 013 104 063 039

PL02 083 192 256 454 051 0005 040 085 107 022

PL03 065 240 239 366 128 0031 025 064 069 044

PL03 115 169 376 361 126 0523 063 142 096 062

PL06 056 175 309 329 165 0012 030 113 085 078

PL07 075 186 412 372 107 0032 038 142 090 048

PL08 043 103 253 222 051 0001 038 158 097 041

PL10 170 257 471 471 218 0031 061 117 082 070

PL11 101 191 289 401 081 0020 049 097 095 035

PL12 160 283 608 404 093 0033 053 138 064 027

PL13 158 190 316 349 110 0124 077 107 083 048

PL14 047 151 298 202 042 0008 029 127 060 023

PL15 074 211 350 238 083 0017 033 106 051 032

PL16 073 292 462 424 183 0065 023 101 065 052

PL16 115 202 510 428 175 0011 053 161 095 072

PL17 069 211 420 447 101 0015 030 128 095 040

PL18 062 218 299 176 033 0020 026 088 036 012

PL18 102 207 356 165 032 0410 046 110 036 013

PL19 104 203 349 444 098 0030 048 110 099 040

PL19-R 083 194 342 450 100 0036 040 113 105 043

PL23 111 237 481 440 163 0011 043 130 084 057

SW01 028 066 078 048 025 0072 039 075 033 031

SW02 299 339 671 676 531 0161 082 127 090 130

SW03 321 399 833 711 556 0100 075 134 080 115

SW04 222 376 699 615 541 0032 055 119 074 119

SW05 004 023 040 011 009 0005 015 111 022 032

SW06 202 227 573 497 290 0077 082 162 099 106

SW07 139 159 237 278 126 0418 081 096 079 066

SW07 138 128 232 282 131 0012 101 117 100 085

SW08 042 077 168 183 112 0030 051 140 107 121

SW09 145 275 499 609 271 0047 049 116 100 081

SW09 233 255 477 583 260 0028 085 120 103 084

SW10 327 362 667 638 470 0224 084 118 079 108

SW11 483 399 813 742 483 0159 113 131 084 100

SW12 415 467 748 849 731 0508 083 103 082 130

SW13 438 381 761 637 499 0268 107 128 075 108

SW14 388 329 731 663 576 0262 109 142 091 145

SW15 451 379 835 745 540 0248 111 141 089 118

SW16 869 740 147 130 893 0263 109 127 079 100

SW17 353 256 563 614 421 0113 128 141 108 136

SW18 483 394 849 101 102 0415 114 138 115 214

SW18 438 377 791 100 105 0362 108 134 120 231

SW19 250 223 522 552 310 0083 104 150 111 115

SW20 412 334 761 550 404 0200 114 146 074 100

SW21 098 091 180 137 146 0332 100 127 068 133

SW22 056 041 185 072 058 0025 128 290 079 117

All the concentration data are blank-corrected subscript N indicates chondrite normalized after McDonough amp Sun(1995)Measured in UMD

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1882

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 27: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig 9 PGE thornRe patterns of the Mengyin (a) Penglai (b) and Shanwang (c) peridotites normalized to the chondrite values of McDonough ampSun (1995) (d^f) PGE ratios R replicate analysis Measured at UMD In (c) dashed lines are for wehrlites SW01 and Fe-rich lherzoliteSW05 from Shanwang that show evidence of metasomatism In (d^f) open diamonds show wehrlites SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios

CHU et al EASTERN NCC MANTLE THINING

1883

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 28: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

of Re Consequently Os model ages for mantle melt resi-dues are more commonly usedTime of Re depletion (TRD) model ages calculated by

assuming that a single melting event removes all of the Refrom a peridotite can provide minimum age constraintsfor mantle melt depletion (Walker et al 1989 Shirey ampWalker 1998) TRD ages will significantly underestimatethe age of a melt depletion event for samples thathave experienced only partial Re removal such as inthe case of lower extents of partial melting and are appli-cable for only very highly refractory peridotites (egAl2O312wt ) ModelTMA ages may be more accurate(Shirey amp Walker 1998) for peridotites that have under-gone only modest melt depletion They are calculatedbased on measured 187Re188Os and record the intersectionof the isotopic evolution of a sample with a mantle evolu-tion line However the accuracy of TMA model ages maybe compromised by Re addition or loss during the historyof the rock but especially in transit to the surface in typi-cally high-Re volcanic systemsBoth TRD and TMA model age calculations are subject

to some further uncertainties First it is clear that thedepleted mid-ocean ridge basalt (MORB) mantle(DMM) is not isotopically homogeneous with respect toOs based on analyses of abyssal peridotites ophiolitesand oceanic basalts For example most abyssal peridotiteshave 187Os188Os ratios ranging from about 0120 to 0129(eg Snow amp Reisberg 1995) but some can have substan-tially lower 187Os188Os (Harvey et al 2006 C Z Liuet al 2008) Thus even the modern DMM includes perido-tite with ancient melt depletion histories recorded in theirOs isotopic compositions (eg Parkinson et al 1998Brandon et al 2000 Meibom amp Frei 2002 Meibom et al2002 Walker et al 2002 2005 Reisberg et al 2004Bizimis et al 2007 Pearson et al 2007 C Z Liu et al2008) The ancient Os is evidently undisturbed by mantlestirring because the extremely high partition coefficientsfor Os in some trace sulfides and alloy phases inhibit lossor gain of Os on a whole-rock scale (Pearson et al 2004)A single sample with a depleted Os isotopic compositionis therefore not strong evidence for the presence of ancientmantle and instead a case for mantle with an ancientprovenance must sometimes be made via the distributionof 187Os188Os ratios for a sizeable suite of samples(Rudnick amp Walker 2009)The second question related to model melt depletion

ages is the extent to which a sample may have beenaffected by melt^rock reaction processes including meta-somatism Reactions involving melts or fluids with high187Os188Os can potentially elevate the 187Os188Os ratio ofthe affected rock and hence decrease the apparent meltdepletion age However most metasomatic fluids andmelts have low Os contents and are inefficient at modifying187Os188Os in peridotites characterized by high Os

concentrations (Handler et al 1997 Shirey amp Walker1998 Walker et al 2002 Chesley et al 2004 Rudnick ampWalker 2009) Only during situations when the meltfluid^rock ratio is high such as when lherzolites or harz-burgites are transformed into dunite are such changesnoted (eg Becker et al 2001 Bulaquo chl et al 2002)Refertilization can also affect the Os isotopic and PGEcomposition of peridotites (Saal et al 2001 Beyer et al2006 Rudnick amp Walker 2009) If refertilization occursshortly following melt extraction as might be expected ina mantle section that experienced adiabatic melting refer-tilization will have no obvious impact on the Re^Ossystem as no time has elapsed between melt depletionand refertilization (Rudnick amp Walker 2009) When alarge time span separates melt depletion and refertiliza-tion the latter will affect the 187Os188Os of the residualmantle but to a lesser degree than major elements such asCaO and Al2O3 which are strongly enriched in the meltrelative to the residual peridotite In this case refertiliza-tion will produce a strongly curved array on an Al2O3 vs187Os188Os plot as Al2O3 is more strongly affected than187Os188Os (Reisberg amp Lorand 1995 Rudnick amp Walker2009) Moreover as the refertilizing melts are usuallyincompatible element (eg LREE) enriched it is likelythat this signature will be imparted to the refertilizedlithospheric mantle and with time the eNd and eHf of theresidues will become negative

Proterozoic lithospheric mantle beneaththe eastern NCC during the PaleozoicPrevious studies have shown that the lithospheric mantlebeneath the eastern North China craton was dominatedby materials with Archean melt depletion ages during thePaleozoic as revealed by mantle peridotites and chromiteseparates from the Paleozoic Mengyin Fuxian and Tielingkimberlites (Gao et al 2002 Wu et al 2006 Zhang et al2008) Our three new late ArcheanTRD model ages (24^27 Ga) obtained for mantle xenoliths from Mengyin(MY9 MY10 and MY34) are consistent with this priorwork However one Mengyin sample (MY33) has a TRD

age of 16 Ga that is much younger despite a highlyrefractory composition (Table 7 Fig 10a) Gao et al (2002)similarly reported a TRD age of 13 Ga for a refractorywhole-rock sample from Mengyin but its relatively high187Re188Os of 048 made the eruption age correction forthis sample potentially problematic Later Zhang et al(2008) reported a TRD age of 18 Ga for a whole-rocksample collected from the Fuxian kimberlite Several anal-yses of chromites from Fuxian and Tieling also yieldedTRD ages of 16^18 Ga (Wu et al 2006)The relatively high S concentrations and LREE enrich-

ments in the Mengyin peridotites suggest that these sam-ples experienced strong metasomatic enrichment (Table 1Fig 5a) so it may be questioned whether the youngermodel ages record a Proterozoic melt depletion event or

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1884

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 29: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig 10 Osmium isotopic compositions vs Al2O3 (wt ) for peridotites from Mengyin (a) Penglai (b) and Shanwang (c) Other data sourcesMengyinccedilGao et al (2002) and Zhang et al (2008) FuxianccedilGao et al (2002) and Zhang et al (2008) TielingccedilWu et al (2006) Osmium isoto-pic compositions of abyssal peridotites are from Martin (1991) Roy-Barman amp Alle gre (1994) Snow amp Reisberg (1995) Brandon et al (2000)Standish et al (2002) Alard et al (2005) Harvey et al (2006) C Z Liu et al (2008) Primitive Upper Mantle (PM open square) from Meiselet al (2001 for Os) and McDonough amp Sun (1995 for Al2O3) open diamonds in (c) are wehrlite SW01 and Fe-rich lherzolite SW05 fromShanwang that probably experienced melt^rock reaction at high melt^rock ratios diagonal shaded area shows refertilization of Archean peri-dotites (3 Ga 35 ppb Os 03wt Al2O3 and an 187Os188Os of 0106) by recent addition of picritic (upper bound picrite melt with 1ppbOs 10wt Al2O3 and an 187Os188Os of 015) or basaltic melt (lower bound basaltic melt has 50 ppt Os 15wt Al2O3 and an 187Os188Osof 015)

CHU et al EASTERN NCC MANTLE THINING

1885

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 30: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

complex isotopic systematics resulting from overprintingHowever none of the Mengyin xenoliths studied hereshow evidence for strong accompanying enrichment of Re(Fig 8) Moreover the highly PPGE-depleted Mengyinxenoliths (Fig 9a) including sample MY33 reflect PPGEremoval rather than addition consistent with highdegrees of melt extraction without subsequent overprint-ing Furthermore IPGE abundances in these rocks are typ-ical of upper mantle materials Thus the Os contained inthese rocks was probably little affected by the apparentmetasomatic event We conclude that although the rockswere modified by subsequent processes the whole-rockmodel TRD ages may represent the dominant melt deple-tion age(s) of the peridotitesCollectively these data suggest that the Ordovician

lithospheric mantle beneath the eastern NCC was domi-nated by refractory Archean peridotite containing aminor amount of Proterozoic peridotite The Proterozoicmodel ages may provide evidence for minor lithos-pheric mantle formation as a consequence of Paleo-Mesoproterozoic mantle upwelling related to Columbiasupercontinent break-up (Zhai amp Liu 2003 Hou et al2008) The limited number of samples coupled with theirhighly altered state precludes a more robust conclusionregarding the absolute age(s) of melt depletion the pro-portion of the Proterozoic material and its position in themantle lithosphere relative to Archean peridotite

Effects of secondary processes onCenozoic mantle xenolithsThe Penglai and many of the Shanwang peridotites appearto have experienced metasomatic overprinting followingmelt depletion as reflected by their REE patterns (Fig 5band c) Moreover there are discrepancies in the mass bal-ance of Sr (and Nd) between whole-rocks and cpx in theShanwang and Penglai peridotites (especially Shanwang)Some samples (eg PL11 PL17 PL18 SW02 SW03 SW04SW10 SW11 SW12 SW19 and SW20) even have higherbulk-rock Sr concentrations than the cpx separates(Tables 1 and 4) As suggested by Rudnick et al (2004) thismay reflect one or both of the following (1) addition of Sr(and Nd) to grain boundaries during metasomatism hostbasalt infiltration or post-eruption alteration (2) the pres-ence of Sr- (and Nd)-bearing accessory phases in theShanwang and Penglai peridotites The discrepancies in Srmass balance in the Shanwang peridotites are greaterthan those of the Penglai samples reflecting stronger Sr-rich meltfluid metasomatism in the Shanwang xenolithsconsistent with their Sr-rich nature (Fig 5f) Actuallysome basalt glass or minor carbonate was identified ongrain boundaries in some samples (see Appendix A)Some unleached cpx with slightly higher 87Sr86Sr andlower 143Nd144Nd than their paired leached cpx may beexplained by a grain boundary phase that is isotopicallyslightly more enriched than the cpx (Song amp Frey 1989

Rudnick et al 2004) Except for sample PL03 unleachedcpx have much higher Rb concentrations than theirpaired leached cpx This suggests that Rb is presentmainly as a grain boundary phase in the peridotite xeno-liths The SmNd ratios calculated from the whole-rockREE data are lower than those of their paired cpx formany Penglai and Shanwang xenoliths especially samplesPL12 PL17 and PL18 This may also reflect later-stageLREE-rich secondary phase addition to these whole-rocks The 176Hf177Hf of leached PL19 cpx is slightlyhigher than its paired unleached cpx This may reflectthe presence of less radiogenic Hf on the grain boundariesThe positive correlation on the Sm^Nd isochron plot isalso consistent with a recent mixing event in the litho-spheric mantle under the eastern NCC where an originalLREE-depleted peridotite with positive eNd (eg thorn12) ismixed with a LREE-enriched melt with much lower eNdA rough positive correlation between 1Nd and143Nd144Nd (not shown) also supports this mixinghypothesisThe Penglai xenoliths might have experienced syn- or

post-eruption sulfide breakdown as reflected in their lowOsIr and high CuS (Tables 1 and 8 Fig 9b) (Handleret al 1999 Pearson et al 2004) Collectively the Penglaiand Shanwang data exhibit a rough negative correlationbetween OsIr and whole rock CuS (not shown) The PdIr ratios of the Penglai samples are also low possiblyreflecting a high degree of prior melt extraction As sug-gested by J G Liu et al (2008) another possibility is thatthe low PdIr ratios of the Pengali xenoliths are alsorelated to syn- or post-eruption sulfide breakdown as arough positive correlation exists between the OsIr andPdIr ratios for the combined Penglai and Shanwang xeno-lith data (Fig 9f)The higher S concentrations in the Shanwang xenoliths

compared with the Penglai xenoliths together with signifi-cant variations in Re and 187Re188Os (Fig 8) may indicatethat the Re^Os system in some Shanwang xenoliths wasinfluenced by sulfide-bearing melt metasomatism Forexample some Shanwang xenoliths (SW12 and SW21 andespecially samples SW01 and SW05) have low Fo numberslow PGE concentrations superchondritic 187Re188Os andOs isotopic compositions that are consistent with melt^rock reaction at high melt^rock ratios (Rudnick ampWalker 2009 and references therein)

Age of the lithospheric mantle beneath theeastern NCC sampled by magmatismduring the CenozoicMantle xenoliths from Cenozoic basalts in eastern Chinagive a wide range of Os model ages fromPaleoproterozoic to modern (Gao et al 2002 Wu et al2003 2006) The meaning of these model ages is highlydebated (Gao et al 2002 Wu et al 2003 2006 Xu et al

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1886

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 31: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

2008 Zhang et al 2008 2009) Like other mantle xenolithsfrom Cenozoic basalts in the eastern NCC the Penglaiand Shanwang xenoliths also show Os model ages rangingfrom Paleoproterozoic to modernThe Penglai xenoliths (eruption age 5 Ma) may pro-

vide the most robust view of the modern lithosphericmantle beneath the eastern NCC yet available Althoughthe Penglai xenoliths have experienced variable metaso-matic overprinting as reflected by their REE and traceelement characteristics their PGE patterns do not showPPGE enrichment Consequently it is likely that the Re^Os isotope systematics of the Penglai xenoliths were notaffected by metasomatic modification Similarly althoughthe Penglai xenoliths have experienced post- or syn-eruption sulfide breakdown as reflected in their low OsIr and high CuS this should not affect their Os isotopicsystematics The Os isotopic compositions of the Penglaixenoliths display a crude positive correlation withAl2O3wt (Fig 10b) Model ages of relatively refractorysamples harzburgites PL08 and PL23 and cpx-poor lher-zolite PL12 are similar at 1 Ga The most refractorysample from this locale (based on its low Al2O3 and CaOcontents) harzburgite PL02 which has an olivine Fo-number of 908 a Cr-number of 394 as well as thelowest HREE concentration in the suite also has a TRD

age of 11 Ga These model ages are in good agreementwith the cpx Lu^Hf errorchron age (13 Ga Fig 6c)obtained from the same suite of samplesEven though there is a fairly close correspondence

between the maximum Os and Hf model ages for thePenglai xenolith suite it should be noted that the two iso-tope systems do not provide concordant results for singlesamples For example the two Penglai samples that definethe Hf errorchron age by having very radiogenic Hf(PL11 and PL19) indicating long-term LREE (and Hf)depletion have 187Os188Os isotope ratios typical of themodern convecting upper mantle (ie 012499 and012597 with corresponding TRD ages of 03^02 Garespectively) (Tables 6 and 7)The reason for these age dis-cordancies for single samples is unclear but probablyrelates to the very different behavior of lithophile andhighly siderophile elements in the upper mantleThe Shanwang xenoliths located within the Tan^Lu

Fault are also typical examples of the young lithosphericmantle beneath the NCC during the Cenozoic (Zhanget al 2006 Zheng et al 2007) Although the Shanwangsamples have experienced some sulfide metasomatism asmentioned above most samples have PUM-like PGEthornRecharacteristics and high Os concentrations (except forSW01 and SW05) implying that their Re^Os isotopeshave not been strongly affected (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009) TheOs isotopic ratios of the Shanwang xenoliths show a crudenegative correlation with the Fo-numbers of the olivine in

the samples (not shown) and a crude positive correlationwith the Al2O3wt of the bulk samples (Fig 10c) Thereis also a rough positive correlation between Os isotopicratios and (PdIr)N (not shown) Collectively these char-acteristics suggest that the different Os isotopic ratiosrelate to the fertility of the xenoliths The model ages ofrelatively refractory Shanwang samples range from 08to 10 GaAlthough mid-Proterozoic OsTRD ages are obtained for

refractory xenoliths from both Penglai and Shanwang itcannot be immediately concluded that this mantle litho-sphere formed in the mid-Proterozoic The modernDMM as recorded by abyssal peridotites and youngophiolites includes peridotites with ancient melt depletionhistories (back to 20 Ga) recorded in their Os isotopiccompositions As suggested by Rudnick amp Walker (2009)for post-Archean peridotites it is necessary to make statis-tical comparisons via measurement of 187Os188Os ratiosfor a sizeable suite of samples to assess whether or not thexenoliths are like the DMM or record an older averagemelt depletionAs shown in Fig 11 the overall distribution of

187Os188Os for the Penglai and Shanwang samples aswell as for other Cenozoic mantle xenoliths from the east-ern NCC is similar to that of the modern convectingmantle Their isotopic compositions are very differentfrom the Paleozoic xenoliths from the eastern NCC(Fig 11) Combined with the fact that the Sr^Nd^Hf iso-tope compositions of the Cenozoic mantle xenoliths fromthe eastern NCC are mostly similar to the modern convect-ing mantle (Fig 7) it is likely that the present lithosphericmantle beneath the eastern NCC is juvenile and formedduring the Mesozoic or CenozoicRecent analyses suggest that materials in the suboceanic

mantle can also retain a long-term Hf isotopic depletionhistory (Salters amp Zindler 1995 Andres et al 2004Graham et al 2006 Bizimis et al 2007) It is possible thatthe convecting mantle contains materials with high Hf iso-topic ratios that never were homogenized as is surmisedfor Os In contrast all data for the Sm^Nd isotope systemsuggest that it is more efficiently homogenized supportingthe conclusion that the Lu^Hf isotope system has a higherclosure temperature than the Sm^Nd and Rb^Sr systems(eg Bedini et al 2004Wittig et al 2006 2007) Thereforealthough a cpx Lu^Hf errorchron of 13 Ga was obtainedthe Lu^Hf results provide permissive evidence that thelithospheric mantle currently underlying the eastern por-tion of the NCC was formed during the Mesozoic orCenozoic from the convective upper mantleThe conclusion that the present lithospheric mantle

beneath the eastern NCC is juvenile may be consistentwith the regional geology Neodymium model ages thatare mostly between 25 and 29 Ga indicate that the NCCwas formed during the Archean (Wu et al 2005b)

CHU et al EASTERN NCC MANTLE THINING

1887

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 32: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

At 185 Ga the NCC underwent cratonization by amal-gamation of the eastern and western blocks (Zhao et al2005) and the crust experienced reworking without anyjuvenile addition The NCC remained stable during theProterozoic and Paleozoic although very minor amountsof anorogenic rapakivi granite and anorthosite have beenidentified locally (Ramo et al 1995 Yang et al 2005 S H

Zhang et al 2007) During the late Paleozoic the northernmargin of the NCC served as an active continentalmargin characterized by subduction of the Paleo-AsianOcean to the north (S H Zhang et al 2009) It was notuntil the Mesozoic that extensive igneous activity occurredthroughout the NCC with significant growth of juvenilecrust (Wu et al 2005a Yang et al 2008) The addition of

Fig 11 Comparison of Os isotopic compositions in modern abyssal peridotites and mantle xenoliths from the North China Craton (a) Abyssalperidotites (data sources as in Fig 10) (b) eastern North China Craton in the Cenozoic (Gao et al 2002Wu et al 2003 2006 this study) and(c) eastern North China Craton in the Early Paleozoic (Gao et al 2002Wu et al 2006 Zhang et al 2008 this study)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1888

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 33: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

juvenile lithospheric mantle beneath the NCC during theMesozoic may thus have accompanied the creation ofnew crust (Wu et al 2005a)

Constraints on the mechanism oflithospheric thinning beneath the NCCIt is widely accepted that a significant part of the deeplithosphere beneath the eastern NCC was removed duringthe Mesozoic However the time extent mechanism andgeodynamic setting of the thinning of the lithosphericmantle underlying the eastern NCC is still highly debated(Menzies et al 2007) Several mechanisms have been pro-posed to explain the lithospheric thinning including dela-mination (eg Gao et al 2002 2004 2008 Wu et al2005a 2007) thermo-mechanical erosion (eg Xu 2001Zheng et al 2007) and lithospheric transformation throughperidotite^melt interaction or chemical erosion (egZhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008)If thermo-mechanical erosion of the lithosphere induced

by the upwelling of the asthenosphere took place graduallyfrom the base of the lithosphere a stratified lithosphericmantle can be envisaged with an Archean relict litho-sphere overlying newly accreted lithosphere (Griffin et al1998 Menzies amp Xu 1998) If asthenospheric upwellingpreferentially focused along weak shear zones in theArchean root (eg Zheng et al 2007) then the Archeanlithospheric mantle would be dispersed in the newlyaccreted Phanerozoic lithospheric mantle However thelithospheric mantle sampled during the Paleozoic appearsto have been mostly Archean (Gao et al 2002 Wu et al2006 Zhang et al 2008 this study) whereas in theCenozoic the mantle lithosphere appears to be exclusivelyjuvenile with no Archean material identified amongst thexenoliths studied to date (Gao et al 2002 Wu et al 20032006 this study) The observation that both garnet- andspinel-facies mantle peridotite xenoliths from theMengyin and Fuxian kimberlites have Archean Re^Os iso-topic ages indicates no chemical stratification of theArchean lithospheric mantle beneath the eastern NCC(Zhang et al 2008) Furthermore recent studies indicatethat peridotite xenoliths hosted by the Mesozoic (125Ma) Laiwu diorites are refractory have low equilibrationtemperatures and have enriched Sr^Nd isotopic composi-tions similar to peridotites from Mengyin (W L Xuet al 2008) The lack of garnet in the xenoliths indicatesthat they were derived from depths580 km Osmium iso-topic compositions indicate that the Laiwu xenoliths areArchean (Gao et al 2008) similar to those fromMengyinThese results further confirm that the refractoryPaleozoic lithospheric mantle beneath the eastern NCC isnot stratified We conclude that the Cenozoic lithosphericmantle is not the upper part of the Paleozoic lithosphericmantle or a relict after thinning

A key question remains as to whether peridotite^meltinteraction or melt addition converted the pre-existingrefractory lithospheric mantle beneath the eastern NCCinto fertile mantle as suggested by recent studies callingupon lithospheric transformation or refertilization(Zhang 2005 Zhang et al 2006 2008 2009 Ying et al2006 Tang et al 2008) However as stated above becausemost metasomatic fluids and melts have Os contents twoto three orders of magnitude lower than those of the peri-dotites it is seemingly difficult to transform peridotiteswith Archean Os model ages to peridotites withProterozoic or Phanerozoic model ages through recentmelt addition (see also Rudnick amp Walker 2009)Furthermore if refertilization occurred much later thanthe formation of the peridotites it should be recognizableon a plot of Al2O3 vs 187Os188Os as a curved trend Asshown in Fig 10 the Os isotopic ratios display a crude pos-itive correlation with Al2O3 for both the Penglai andShanwang xenoliths and lie far above the shaded areawithin which they would plot if they represented Archeanlithosphere (3 Ga 35 ppb Os 03wt Al2O3 and an187Os188Os of 0106) that had experienced recent meltaddition (upper bound picrite melt with 1ppb Os 10wt Al2O3 and an 187Os188Os of 015 lower bound basalticmelt with 50 ppt Os 15 wt Al2O3 and an 187Os188Os of015) Collectively the data do not support the idea thatthe Cenozoic peridotites are Archean peridotites that haveexperienced significant melt addition in the Mesozoic orlater As argued above the Shanwang xenoliths may haveexperienced intensive sulfide-bearing melt metasomatismHowever even if sulfides were precipitated the metasoma-tized mantle should show significant enrichment of PPGEalong with high ReOs ratios (Pearson et al 2004Reisberg et al 2004 2005 Rudnick amp Walker 2009)which are not observed in most Shanwang xenolithsMoreover secondary sulfides typically have one to twoorders of magnitude lower Os content than primary sul-fides (Alard et al 2000) which means it may also be diffi-cult to transform peridotites with Archean Os model agesto peridotites with Proterozoic or Phanerozoic model agesthrough recent sulfide metasomatism (Gao et al 2002Reisberg et al 2004 2005) Collectively the present-daylithospheric mantle beneath the eastern NCC is probablynot refertilized or metasomatized Archean lithosphericmantleIt has been long recognized that the mantle lithosphere

that existed beneath the eastern NCC during thePaleozoic was highly refractory with an enriched Sr^Ndisotope composition in contrast to the lithospheric mantlethat underlies this region at present which is fertile andhas a Sr^Nd isotope composition similar to that of thedepleted mantle (eg Griffin et al 1998 Menzies et al2007) as shown in the present study (Fig 7) We have alsoshown on the basis of Os isotopes that the Archean

CHU et al EASTERN NCC MANTLE THINING

1889

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 34: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

lithosphere was replaced after the Ordovician by juvenilelithospheric mantle that has Os isotopic characteristicsindistinguishable from the modern convecting uppermantle (Fig 11) Overall the present lithospheric mantlebeneath the eastern NCC is completely different from thatpresent in the Paleozoic and is much younger than itsoverlying Archean crust It is possible that the ancientlithospheric mantle beneath the eastern NCC was removedby foundering during the Mesozoic or later and replacedby the cooled product of upwelling asthenospheric mantleCombined with other lines of evidence from Mesozoicmagmatic rocks (eg Gao et al 2004 2008 Wu et al2005a) the delamination model is our preferred mecha-nism to account for the loss of the ancient keel that onceunderlay the North China Craton

CONCLUSIONS(1) Archean Re^Os model ages for peridotite xenolithshosted by the Paleozoic Mengyin diamondiferous kimber-lites provide further evidence that Archean lithosphericmantle dominated the eastern North China Craton litho-sphere during the Paleozoic Our new data also indicatethat minor Paleoproterozoic lithosphere was also present(2) Like peridotite xenoliths from other Cenozoic basalts

in the eastern North China Craton the Shanwang andPenglai xenoliths have Os model ages that range back tothe Mesoproterozoic The distribution of Os isotopic com-positions for these xenoliths however is identical to thatseen in modern abyssal peridotites which sample the pre-sent-day convecting mantle This similarity betweenCenozoic lithosphere and abyssal peridotites suggests thatthis lithosphere formed during the Mesozoic or Cenozoicit is unlikely that the present lithospheric mantle beneaththe eastern NCC was refertilized or metasomatizedArchean lithospheric mantle(3) The high initial eNd (5^24 averagefrac1410) and eHf (7^

75 averagefrac14 25 excluding the two very radiogenic sam-ples) and low 87Sr86Sr (07023^07042 averagefrac14 07030)of cpx from the Cenozoic peridotites are consistent withtheir derivation from convecting upper mantle during theMesozoic or Cenozoic(4) Nd^Hf isotope decoupling in cpx from the Penglai

peridotites indicates that the convecting upper mantlemay contain materials with radiogenic Hf isotopic ratiossimilar to the ancient Os that is occasionally recorded inabyssal peridotites whereas Nd isotopes do not reflectsuch ancient melt depletions

ACKNOWLEDGEMENTSYan-Bin Zhang Chao-Feng Li Qian Mao Yu-Guang MaHe Li Jie Cao and Jianfeng Gao are thanked for theirhelp in obtaining trace elemental Sr^Nd isotopic EPMAXRF and PGE data Chuan-zhou Liu Ji-feng Ying Qin

Zhou and Zhi-chao Liu are thanked for their help withthe photomicrography We also thank J Blusztajn HBecker and an anonymous reviewer for their constructivereviews This work was financially supported by NationalNatural Science Foundation of China (NSFC Grants40634019) and the US National Science Foundation (EAR0635671)

SUPPLEMENTARY DATASupplementary data for this paper are available at Journalof Petrology online

REFERENCESAlard O GriffinW L Lorand J P Jackson S E amp OrsquoReilly SY

(2000) Non-chondritic distribution of the highly siderophile ele-ments in mantle sulphides Nature 407 891^894

Alard O Luguet A Pearson N J Griffin W L Lorand J PGannoun A Burton KW amp OrsquoReilly SY (2005) In situ Os iso-topes in abyssal peridotites bridge the isotopic gap betweenMORBs and their source mantle Nature 436 1005^1008

Andres M Blichert-Toft J amp Schilling J G (2004) Nature of thedepleted upper mantle beneath the Atlantic evidence from Hf iso-topes in normal mid-ocean ridge basalts from 798N to 558S Earthand Planetary Science Letters 225 89^103

Becker H Shirey S B amp Carlson RW (2001) Effects of melt perco-lation on the Re^Os systematics of peridotites from a Paleozoicconvergent plate margin Earth and Planetary Science Letters 188107^121

Becker H Carlson R W amp Shirey S B (2004) Slab-derivedosmium and isotopic disequilibrium in garnet pyroxenites from aPaleozoic convergent plate margin (lower Austria) Chemical

Geology 208 141^156Becker H Horan M F Walker R J Gao S Lorand J P amp

Rudnick R L (2006) Highly siderophile element composition ofthe Earthrsquos primitive upper mantle Constraints from new data onperidotite massifs and xenoliths Geochimica et Cosmochimica Acta 704528^4550

Bedini R M Blichert-Toft J Boyet M amp Albare de F (2004)Isotopic constraints on the cooling of the continental lithosphereEarth and Planetary Science Letters 223 99^111

Bertrand P amp Mercier J C (1985) The mutual solubility of coexist-ing ortho- and clinopyroxene toward an absolute geothermometerfor the natural system Earth and Planetary Science Letters 76109^122

Beyer E E GriffinW L amp OrsquoReilly SY (2006) Transformation ofArchean lithospheric mantle by refertilization evidence fromexposed peridotites in theWestern Gneiss Region Norway Journalof Petrology 47 1611^1636

Birck J L Roy-Barman M amp Capmas F (1997) Re^Os isotopicmeasurements at the femtomole level in natural samplesGeostandards Newsletter 20 19^27

Bizimis M Griselin M Lassiter J C Salters V J M amp Sen G(2007) Ancient recycled mantle lithosphere in the Hawaiianplume osmium^hafnium isotopic evidence from peridotite mantlexenoliths Earth and Planetary Science Letters 257 259^273

Boyd F R (1989) Compositional distinction between oceanic andcratonic lithosphere Earth and Planetary Science Letters 96 15^26

Brandon A D Snow J EWalker R J Morgan JW amp Mock TD (2000) 190Pt^186Os and 187Re^187Os systematics of abyssal peri-dotites Earth and Planetary Science Letters 177 319^335

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1890

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 35: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Brey G P amp Kohler T (1990) Geothermobarometry in four-phaselherzolites II new thermobarometers and practical assessment ofexisting thermobarometers Journal of Petrology 31 1353^1378

Bulaquo chl A Brulaquo gmann G Batanova V G Mulaquo nker C ampHofmann AW (2002) Melt percolation monitored by Os isotopesand HSE abundances a case study from the mantle section of theTroodos ophiolite Earth and Planetary Science Letters 204 385^402

Carlson RW Pearson D G amp James DE (2005) Physical chemi-cal and chronological characteristics of continental mantleReview of Geophysics 43 2004RG000156

Chen L Zheng T Y amp Xu W W (2006) A thinned lithosphericimage of the Tanlu Fault Zone eastern China constructed fromwave equation based receiver function migration Journal of

Geophysical Research 111 B09312 doi1010292005JB003974Chesley J Righter K amp Ruiz J (2004) Large-scale mantle metaso-

matism a Re^Os perspective Earth and Planetary Science Letters 21949^60

Chi J S amp Lu F X (1996) Kimberlites and the features of Paleozoic litho-spheric mantle in North China craton Beijing Science Press 292 pp

Chu Z Y Chen F K Wang W Xie L W amp Yang YH (2007)High-precision measurement for the concentration and isotopiccomposition of Rhenium and Osmium in micro-amount of geologi-cal samples (in Chinese with English abstract) Rock and Mineral

Analysis 26 431^435Cohen A S amp Waters F G (1996) Separation of osmium from geo-

logical materials by solvent extraction for analysis by TIMSAnalytica Chimica Acta 332 269^275

Dobbs P N Duncan D J Hu S Shee S R Colgan E ABrown M A Smith C B amp Allsopp H P (1994) The geologyof Mengyin kimberlites Shandong China In Meyer O A ampLeonardos O H (eds) Kimberlites Related Rocks and Mantle

Xenoliths CPRMSpecial Publication 1A93 40^61Dulski P (2001) Reference materials for geochemical studies new

analytical data by ICP-MS and critical discussion of referencevalues Geostandards and Geoanalytical Research 25 87^125

FanW M amp Menzies M A (1992) Destruction of aged lower litho-sphere and accretion of asthenosphere mantle beneath easternChina Geotectonica et Metallogenia 16 171^180

FanW M Zhang H F Baker J Jarvis K E Mason P R D ampMenzies M A (2000) On and off the North China craton whereis the Archean keel Journal of Petrology 41 933^950

Faure M LinW Monie P amp Bruguier O (2004) Paleoproterozoicarc magmatism and collision in Liaodong Peninsula (north-eastChina)Terra Nova 16 75^80

Foley S F (2008) Rejuvenation and erosion of the cratonic litho-sphere Nature Geoscience 1 503^510

Gao S Rudnick R L Carlson RW McDonoughW F amp LiuYS (2002) Re^Os evidence for replacement of ancient mantle litho-sphere beneath the North China Craton Earth and Planetary ScienceLetters 198 307^322

Gao S Rudnick R LYuan H L Liu X M LiuY S XuW LLing W L Ayers J Wang X C amp Wang Q H (2004)Recycling lower continental crust in the North China cratonNature 432 892^897

Gao S Rudnick R L XuW LYuan H L LiuY SWalker RJ Puchtel I Liu X M Huang H Wang X R amp Yang J(2008) Recycling deep cratonic lithosphere and generation of intra-plate magmatism in the North China Craton Earth and Planetary

Science Letters 270 41^53Govindaraju K (1994) 1994 compilation of working values and

sample description for 383 geostandards Geostandards Newsletter

18(Special Issue) 1^158

Graham D W Blichert-Toft J Russo C J Rubin KH ampAlbare de F (2006) Cryptic striations in the upper mantle revealedby hafnium isotopes in southeast Indian ridge basalts Nature 440199^202

Griffin W L Zhang A D OrsquoReilly S Y amp Ryan C G (1998)Phanerozoic evolution of the lithosphere beneath the Sino-Koreancraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 107^126Griffin W L OrsquoReilly S Y amp Ryan C G (1999) The composition

and origin of sub-continental lithospheric mantle In Fei YBertka M amp Mysen B O (eds) Mantle Petrology Field Observations

and High-Pressure Experimentation A Tribute to France R (Joe) Boyd

Geochemical Society Special Publications 6 13^46Handler M R BennettV C amp EsatT Z (1997) The persistence of

off-cratonic lithospheric mantle Os isotopic systematics of variablymetasomatised southeast Australian xenoliths Earth and Planetary

Science Letters 151 61^75Handler M R Bennett V C amp Dreibus G (1999) Evidence from

correlated IrOs and CuS for late-stage Os mobility in peridotitexenoliths Implications for Re^Os systematics Geology 27 75^78

Harvey J Gannoun A Burton KW Rogers NW Alard O ampParkinson I J (2006) Ancient melt extraction from the oceanicupper mantle revealed by Re^Os isotopes in abyssal peridotitesfrom the Mid-Atlantic ridge Earth and Planetary Science Letters 244606^621

HermannW amp Berry R F (2002) MINSQccedila least squares spread-sheet method for calculating mineral proportions from whole rockmajor element analyses Geochemistry Exploration Environment

Analysis 2 361^368Hou G T Li J H YaoW H Wang C C amp Wang Y X (2008)

Geochemical constraints on the tectonic environment of the LatePaleoproterozoic mafic dyke swarms in the North China CratonGondwana Research 13 103^116

Lee CT amp Rudnick R L (1999) Compositionally stratified cratoniclithosphere petrology and geochemistry of peridotite xenolithsfrom the Labait volcano Tanzania In Gurney J J Gurney JL Pascoe M D amp Richardson S R (eds) The P H Nixon

Volume Proceedings of 7th Internation Kimberlite Conference Cape TownRed Roof Design pp 503^521

Liu C Z Snow J E Hellebrand E Brulaquo gmann G von derHandt A Bulaquo chl A amp Hofmann A W (2008) Ancient highlyheterogeneous mantle beneath Gakkel Ridge Arctic OceanNature 452 311^316

Liu D Y Nutman A P Compston W Wu J S amp Shen Q H(1992) Remnants of 43800 Ma crust in the Chinese part of theSino-Korean Craton Geology 20 339^342

Liu J G Rudnick R L Walker R J Gao S amp Wu F-Y (2008)The age of mantle lithosphere on the northernmost margin of theCentral Orogenic Belt North China Craton In Abstracts 9thInternational Kimberlite Conference Frankfurt World Wide WebAddress httpwww9ikcuni-frankfurtdeindexhtml

Liu J Q (1999) Active Volcanoes in China (in Chinese) Beijing SciencePress 219 pp

Lu X PWu FY Guo J HWilde S AYang J H Liu X M ampZhang X O (2006) Zircon U^Pb geochronological constraintson the Paleoproterozoic crustal evolution of the Eastern Block inthe North China Craton Precambrian Research 146 138^164

Makishima A amp Nakamura E (2006) Determination of majorminor and trace elements in silicate samples by ICP-QMS andICP-SFMS applying isotope dilution-internal standardisation(ID-IS) and multi-stage internal standardisation Geostandards andGeoanalytical Research 30 245^271

CHU et al EASTERN NCC MANTLE THINING

1891

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 36: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Martin C E (1991) Osmium isotopic characteristics of mantle-derived rocks Geochimica et Cosmochimica Acta 55 1421^1434

Masuda A Nakamura N amp Tanaka T (1973) Fine structures ofmutually normalised rare-earth patterns of chondrites Geochimicaet Cosmochimica Acta 37 239^244

McDonoughW F amp Frey F A (1989) Rare earth elements in uppermantle rocks In Lipin B R amp McKay G A (eds) Geochemistryand Mineralogy of Rare Earth Elements Mineralogical Society of America

Reviews in Mineralogy 21 99^145McDonoughW F amp Sun S S (1995) The composition of the Earth

Chemical Geology 120 223^253Meibom A amp Frei R (2002) Evidence for an ancient osmium isoto-

pic reservoir in Earth Science 296 516^518Meibom A Sleep N H Chamberlain C P Coleman R G

Frei R Hren M T amp Wooden J L (2002) Re^Os isotopic evi-dence for long-lived heterogeneity and equilibration processes inEarthrsquos upper mantle Nature 419 705^708

Meisel TWalker R J Irving A J amp Lorand J P (2001) Osmiumisotopic compositions of mantle xenoliths a global perspectiveGeochimica et Cosmochimica Acta 65 1311^1323

Menzies M A amp XuY G (1998) Geodynamics of the North ChinaCraton In Flower M F J Chung S L Lo C H amp Lee T Y(eds) Mantle Dynamics and Plate Interactions in East Asia American

Geophysical Union Geodynamic Series 27 155^165Menzies M A Fan W M amp Zhang M (1993) Palaeozoic and

Cenozoic lithoprobe and the loss of 4120 km of Archean litho-sphere Sino-Korean craton China In Prichard H MAlabaster T Harris N B W amp Neary C R (eds) Magmatic

Processes and Plate Tectonics Geological Society London Special

Publications 76 71^81Menzies M A Xu Y G Zhang H F amp Fan W M (2007)

Integration of geology geophysics and geochemistry A key tounderstanding the North China Craton Lithos 96 1^21

Nowell G M Kempton P D Noble S R Fitton J GSaunders A D Mahoney J J amp Taylor R N (1998) High preci-sion Hf isotope measurements of MORB and OIB by thermal ioni-zation mass spectrometry Insights into the depleted mantleChemical Geology 149 211^233

Parkinson I J Hawkesworth C J amp Cohen A S (1998) Ancientmantle in a modern arc osmium isotopes in Izu^Bonin^Marianaforearc peridotites Science 281 2011^2013

Pearson D G amp Woodland S J (2000) Solvent extractionanionexchange separation and determination of PGEs (Os Ir Pt PdRu) and Re^Os isotopes in geological samples by isotope dilutionICP-MS Chemical Geology 165 87^107

Pearson D G Irvine G J Ionov D A Boyd F R amp Dreibus GE (2004) Re^Os isotope systematics and platinum group elementfractionation during mantle melt extraction a study of massif andxenolith peridotite suites Chemical Geology 208 29^59

Pearson D G Parman SW amp Nowell G M (2007) A link betweenlarge mantle melting events and continent growth seen in osmiumisotopes Nature 449 202^205

Qi L Zhou M F Malpas J amp Sun M (2005) Determination ofrare earth elements andY in ultramafic rocks by ICP-MS after pre-concentration using Fe(OH)3 and Mg(OH)2 coprecipitationGeostandards and Geoanalytical Research 29 131^141

Ramo O T Haapala I Vaasjoki M Yu J H amp Fu H Q (1995)1700 Ma Shachang complex northeast China Proterozoic rapakivigranite not associated with Paleoproterozoic orogenic crustGeology 23 815^818

Rehkalaquo mper M Halliday A N Barford D Fitton J G ampDawson J B (1997) Platinum-group element abundance patternsin different mantle environments Science 278 1595^1598

Reisberg L amp Lorand J P (1995) Longevity of sub-continentalmantle lithosphere from osmium isotope systematics in orogenicperidotite massifs Nature 376 159^162

Reisberg L Lorand J P amp Bedini R M (2004) Reliability of Osmodel ages in pervasively metasomatized continental mantle litho-sphere a case study of Sidamo spinel peridotite xenoliths (EastAfrican Rift Ethiopia) Chemical Geology 208 119^140

Reisberg L Zhi X C Lorand J P Wagner C Peng Z C ampZimmermann C (2005) Re^Os and S systematics of spinel perido-tite xenoliths from east central China Evidence for contrastingeffects of melt percolation Earth and Planetary Science Letters 239286^308

Roy-Barman M amp Alle gre C J (1994) 187Os186Os ratios of mido-cean ridge basalts and abyssal peridotites Geochimica et

Cosmochimica Acta 58 5043^5054Rudnick R L amp Walker R J (2009) Interpreting ages from Re^Os

isotopes in peridotites In Brey G P Aulbach S Foley SGrulaquo tter H S Holaquo fer H Jacob D E Lorenz V Stachel T ampWoodland A B (eds) Proceedings of the Ninth International KimberliteConference Lithos doi101016jlithos200904042

Rudnick R L Gao S LingW L LiuY S amp McDonoughW F(2004) Petrology and geochemistry of spinel peridotite xenolithsfrom Hannuoba and Qixia North China craton In Mitchell RH Grulaquo tter H S Heaman L M Scott Smith B H ampStachel T (eds) Proceedings of the 8th International Kimberlite

Conference Lithos 77 609^637Saal A E Takazawa E Frey F A Shimizu N amp Hart S R

(2001) Re^Os isotopes in the Horoman peridotite evidence forrefertilization Journal of Petrology 42 25^37

Salters V J amp Zindler A (1995) Extreme 176Hf177Hf in the sub-oceanic mantle Earth and Planetary Science Letters 129 13^30

Shi S Y Wen H L Li B He H L amp Lu C F (2001)Determination of carbon and sulfur in geological samples by highfrequency IR-absorption spectrometric method (in Chinese withEnglish abstract) Rock and Mineral Analysis 20 267^271

Shirey S B amp Walker R J (1995) Carius tube digestions for low-blank rhenium^osmium analysis Analytical Chemistry 67 2136^2141

Shirey S B amp Walker R J (1998) The Re^Os isotope system in cos-mochemistry and high temperature geochemistry Annual Review of

Earth and Planetary Sciences 26 423^500Snow J E amp Reisberg L (1995) Os isotopic systematics of the

MORB mantle results from altered abyssal peridotites Earth and

Planetary Science Letters 136 723^733Song Y amp Frey F A (1989) Geochemistry of peridotite xenoliths in

basalt from Hannuoba Eastern China implications for subconti-nental mantle heterogeneity Geochimica et Cosmochimica Acta 5397^113

Standish J J Hart S R Blusztajn J Dick H J B amp Lee K L(2002) Abyssal peridotite osmium isotopic compositions from Cr-spinel Geochemistry Geophysics Geosystems 3 1004 doi1010292001GC000161

Tang Y J Zhang H F Ying J F Zhang J amp Liu X M (2008)Refertilization of ancient lithospheric mantle beneath the centralNorth China Craton Evidence from petrology and geochemistryof peridotite xenoliths Lithos 101 435^452

Vervoort J D Patchett P J Blichert-Toft J amp Albare de F (1999)Relationships between Lu^Hf and Sm^Nd isotopic systems in theglobal sedimentary system Earth and Planetary Science Letters 16879^99

Walker R J Carlson RW Shirey S B amp Boyd F R (1989) Os SrNd and Pb isotope systematics of southern African peridotitexenoliths Implications for the chemical evolution of subcontinentalmantle Geochimica et Cosmochimica Acta 53 1583^1595

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1892

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 37: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Walker R J Prichard H M Ishiwatari A amp Pimentel M (2002)The osmium isotopic composition of convecting upper mantlededuced from ophiolite chromites Geochimica et Cosmochimica Acta

66 329^345Walker R J Brandon A D Bird J M Piccoli P M

McDonoughW F amp Ash R D (2005) 187Os^186Os systematics ofOs^Ir^Ru alloy grains from southwestern Oregon Earth and

Planetary Science Letters 230 211^226Walker R J McDonough W F Honesto J Chabot N L

McCoy T J Ash R D amp Bellucci J J (2008) Modeling frac-tional crystallization of group IVB iron meteorites Geochimica et

Cosmochimica Acta 72 2198^2216WangWYTakahashi E amp Sueno S (1998) Geochemical properties

of lithospheric mantle beneath the Sino-Korea craton evidencefrom garnet xenocrysts and diamond inclusions Physics of the Earthand Planetary Interiors 107 249^260

Wells P R A (1977) Pyroxene thermometry in simple and complexsystems Contributions to Mineralogy and Petrology 62 129^139

Wittig N Baker J A amp Downes H (2006) Dating the mantle rootsof young continental crust Geology 34 237^240

Wittig N Baker J A amp Downes H (2007) U^Th^Pb and Lu^Hfisotopic constraints on the evolution of sub-continental lithosphericmantle French Massif Central Geochimica et Cosmochimica Acta 711290^1311

Wood B J amp Banno S (1973) Garnet^orthopyroxene and orthopyr-oxene^clinopyroxene relationships in simple and complex systemsContributions to Mineralogy and Petrology 42 109^124

Wu FYWalker R J Ren XW Sun DY amp Zhou X H (2003)Osmium isotopic constraints on the age of lithospheric mantlebeneath northeastern China Chemical Geology 197 107^129

Wu FY Lin J QWilde S A Zhang X O amp Yang JH (2005a)Nature and significance of the Early Cretaceous giant igneousevent in Eastern China Earth and Planetary Science Letters 233103^119

Wu FY Zhao G CWilde S A amp Sun DY (2005b) Nd isotopicconstraints on the crustal formation of the North China CratonJournal of Asian Earth Sciences 24 523^545

Wu F Y Yang J H Wilde S A amp Zhang XO (2005c)Geochronology petrogenesis and tectonic implications of Jurassicgranites in the Liaodong Peninsula NE China Chemical Geology221 127^156

Wu FYWalker R JYangY HYuan H L amp Yang J H (2006)The chemical^temporal evolution of lithospheric mantle underly-ing the North China Craton Geochimica et Cosmochimica Acta 705013^5034

Wu F Y Yang J H Chu Z Y Xie LW Yang Y H amp Li Q L(2007) Dating the subcontinental lithospheric mantle (in Chinesewith English abstract) Earth Science Frontiers 14 76^86

Wu F Y Zhang Y B Yang J H Xie L W amp Yang Y H (2008)Zircon U^Pb and Hf isotopic constraints on the Early Archeancrustal evolution in Anshan of the North China CratonPrecambrian Research 167 339^362

Xu W L Hergt J M Gao S Pei F P Wang W amp Yang D P(2008) Interaction of adakitic melt^peridotite implications for thehigh^Mg signature of Mesozoic adakitic rocks in the easternNorth China Craton Earth and Planetary Science Letters 265 123^137

Xu X S GriffinW L OrsquoReilly SY Pearson N J Geng HY ampZheng J P (2008a) Re^Os isotopes of sulfides in mantle xenolithsfrom eastern China Progressive modification of lithosphericmantle Lithos 102 43^64

Xu Y G (2001) Thermo-tectonic destruction of the Archaean litho-spheric keel beneath eastern China evidence timing and mechan-ism Physics and Chemistry of the Earth (A) 26 747^757

Yang J H Wu F Y Liu X M amp Xie LW (2005) Zircon U^Pbages and Hf isotopes and their geological significance of theMiyun rapakivi granites from Beijing China Acta Petrologica Sinica21 1633^1644

Yang J H Wu F Y Wilde S A Belousova E amp Griffin W L(2008) Mesozoic decratonization of the North China block Geology36 467^470

YangY HWu FYWilde S A Liu X M ZhangY B Xie LW amp Yang J H (2009) In-situ perovskite Sr^Nd isotopic con-straints on petrogenesis of the Mengyin kimberlites in the NorthChina Craton Chemical Geology 264 24^42

Ying J F Zhang H F Kita N Morishita Y amp Shimoda G(2006) Nature and evolution of Late Cretaceous lithosphericmantle beneath the eastern North China Craton Constraintsfrom petrology and geochemistry of peridotitic xenoliths fromJulaquo nan Shandong Province China Earth and Planetary Science

Letters 244 622^638Zhai M G amp Liu W J (2003) Paleoproterozoic tectonic history of

the North China Craton a review Precambrian Research 122183^199

Zhang H F (2005) Transformation of lithospheric mantle throughperidotite^melt reaction A case of Sino-Korean craton Earth and

Planetary Science Letters 237 768^780Zhang H F amp YangY H (2007) Emplacement age and Sr^Nd^Hf

isotopic characteristics of the diamondiferous kimberlites from theeastern North China Craton (in Chinese with English abstract)Acta Petrologica Sinica 23 285^294

Zhang H FYing J FTangY J Zhang J Zhao X M Niu L FXiao Y amp Su B X (2006) Heterogeneity of Mesozoic andCenozoic lithospheric mantle beneath the eastern North ChinaCraton evidence from olivine compositional mapping (in Chinesewith English abstract) Acta Petrologica Sinica 22 2279^2288

Zhang H F Goldstein S L Zhou X H Sun M Zheng J P ampCai Y (2008) Evolution of subcontinental lithospheric mantlebeneath eastern China Re^Os isotopic evidence from mantlexenoliths in Paleozoic kimberlites and Mesozoic basaltsContributions to Mineralogy and Petrology 155 271^293

Zhang H F Goldstein S L Zhou X H Sun M amp CaiY (2009)Comprehensive refertilization of lithospheric mantle beneath theNorth China Craton further Os^Sr^Nd isotopic constraintsJournal of the Geological Society London 166 249^259

Zhang S H Liu SW ZhaoYYang J H Song B amp Liu X M(2007) The 175^168 Ga anorthosite^mangerite^alkali granitoid^rapakivi granite suite from the northern North China Cratonmagmatism related to a Paleoproterozoic orogen Precambrian

Research 155 287^312Zhang S H Zhao Y Song B Hu J M Liu S W Yang Y H

Chen F K Liu X M amp Liu J (2009) Contrasting LateCarboniferous and Late Permian^Middle Triassic intrusive suitesfrom the northern margin of the North China cratonGeochronology petrogenesis and tectonic implications GeologicalSociety of America Bulletin 121 181^200

Zhao G C Sun MWilde S A amp Li S Z (2005) Late Archean toPaleoproterozoic evolution of the North China Craton key issuesrevisited Precambrian Research 136 177^202

Zheng J P (1999) Mesozoic^Cenozoic mantle replacement and lithospheric

thinning beneath the eastern China (in Chinese with English abstract)Wuhan China University of Geosciences Press 126 pp

Zheng J P Sun M Zhou M F amp Robinson P (2005) Trace ele-mental and PGE geochemical constraints of Mesozoic andCenozoic peridotitic xenoliths on lithospheric evolution of theNorth China Craton Geochimica et Cosmochimica Acta 69 3401^3418

CHU et al EASTERN NCC MANTLE THINING

1893

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 38: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Zheng J P Griffin W L OrsquoReilly S Y Yang J S Li T FZhang M Zhang R Y amp Liou J G (2006) Mineral chemistryof peridotites from Paleozoic Mesozoic and Cenozoic lithosphereconstraints on mantle evolution beneath eastern China Journal ofPetrology 47 2233^2256

Zheng J PYu C M Lu F X Zhang Z H amp Tang HY (2007)Age and composition of continental mantle peridotites and impli-cations for the lithospheric thinning eastern North China (inChinese with English abstract) Earth Science Frontiers 14 87^97

APPENDIX A PETROGRAPH IC DESCR IPT IONS OF THE ANALYZEDPER IDOTITES

Sample Rock type Mineral assemblage Description

(modal )

PL01 Lherzolite Ol (59)thornOpx (26)thornCpx (12)thornSp (3) Porphyroclastic texture with 3ndash5mm grain-size of relict exsolved Opx

The groundmass has a grain-size of 1ndash2mm with sieve-textured Cpx

PL02 Harzburgite Ol ((75)thornOpx (21)thornCpx (3)thornSp (1) Protogranular texture with grain-size of 1ndash2mm Ol Opx and Cpx show equilibrium

grain boundaries Some spinels have broken down with variable residues

PL03 Lherzolite Ol (53)thornOpx (30)thornCpx (14)thornSp (3) Protogranular texture with grain-size of 1mm Most Cpx show reaction rims

PL06 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (4) Coarse-grained equigranular texture with grain-size of 2ndash3mm Ol Opx and Cpx

form triple junctions Some Cpx are modified with sieve-textured rims

PL07 Lherzolite Ol (56)thornOpx (25)thornCpx (15)thornSp (3) Protogranular texture with grain-size of 1mm Opx porphyoclasts can be found

occasionally and no reaction rim is identified for Cpx

PL08 Lherzolite Ol (73)thornOpx (19)thornCpx (6)thornSp (2) Porphyroclastic texture with 3mm Opx blasts and 1mm matrix crystals

Ol occurs as tabular crystals

PL13 Lherzolite Ol (59)thornOpx (25)thornCpx (13)thornSp (3) Coarse-grained peridotite with 4ndash6mm porphyroclastic Opx the rest of the miner-

als are 3mm

PL14 Lherzolite Ol (61)thornOpx (23)thornCpx (13)thornSp (3) Porphyroclastic texture with 3mm Opx The groundmass is51mm

PL16 Lherzolite Ol (58)thornOpx (25)thornCpx (14)thornSp (3) Medium-grained equigranular texture with a grain-size of 1ndash2mm triple junctions

are well developed between OlCpx and Opx

PL17 Lherzolite Ol (60)thornOpx (26)thornCpx (11)thornSp (3) Medium-grained equigranular texture with grain-size of 1ndash2mm Ol Opx and Cpx

mostly form equilibrated triple junctions Some Cpx develops sieve texture along

the grain margin

PL18 Harzburgite Ol (71)thornOpx (22)thornCpx (5)thornSp (3) Tabular equigranular texture with grain-size of 1mm Most Sp is decomposed

with a reaction rim

PL19 Lherzolite Ol (68)thornOpx (22)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 1mm Most Ol Opx and Cpx form triple

junctions indicating their equilibrium occurrence Most Sp have reaction rim

PL23 Harzburgite Ol (64)thornOpx (30)thornCpx (4)thornSp (2) Coarse-grained protogranular texture with grain-size o f 3ndash5mm Cpx is character-

ized by reaction rim

SW01 Wehrlite Resorption texture with grain-size of 3ndash5mm Ol is surrounded by sieve-textured

clinopyroxene Some Ol are weakly serpentinized

SW02 Lherzolite Ol (59)thornOpx (21)thornCpx (18)thornSp (3) Protogranular texture with grain-size of 05ndash15mm Opx porphyroclasts occasion-

ally occur up to 4mm Serpentinization is observed on fractures along with

minor carbonates

SW03 Lherzolite Ol (52)thornOpx (35)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1mm Most Ol are serpentinized with few

fresh grains

SW04 Pyroxenite Ol (35)thornOpx (15)thornCpx (50)thornSp (51) Mosaic porphyroclastic texture Opx forms porphyroclasts with maximum grain-size

of 5mm The groundmass composed of Ol Cpx and Opx is usually504mm

Basaltic glass and minor carbonates are identified on grain boundaries

SW05 Lherzolite Ol (64)thornOpx (23)thornCpx (11)thornSp (2) Protogranular texture with grain-size of 1ndash2mm Ol are weakly serpentinized Cpx

is sieve-textured Sp mostly occur within the sieve-textured Cpx as inclusions

(continued)

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1894

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 39: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Appendix A Continued

Sample Rock type Mineral assemblage Description

(modal )

SW06 Lherzolite Ol (61)thornOpx (27)thornCpx (10)thornSp (2) Porphyroclastic texture Opx porphyroclasts have a grain size of 2mm Other

minerals are505mm

SW07 Harzburgite Ol (77)thornOpx (18)thornCpx (5)thornSp (1) Equigranular texture with banded structure Cpx mostly has thin reaction rims and

Sp occurs along the layering

SW08 Lherzolite Ol (61)thornOpx (22)thornCpx (14)thornSp (3) Equigranular texture with grain-size of 51mm Some Cpx and Sp show sieve

textures along their margins

SW09 Lherzolite Ol (73)thornOpx (18)thornCpx (8)thornSp (2) Equigranular texture with grain-size of 51mm weak serpentinization Most Cpx

show reaction rims

SW10 Lherzolite Ol (63)thornOpx (22)thornCpx (12)thornSp (3) Porphyroclastic texture with Opx and sometimes Ol occurring as porphyroclasts

with grain-size of 4mm Some Ol inclusions in Opx porphyroclasts

SW11 Lherzolite Ol (65)thornOpx (21)thornCpx (13)thornSp (2) Granuloblastic texture with equilibrated Ol Opx and Cpx Serpentinization occurs

along fractures in which very thin (501mm width) carbonate vein is identified

SW12 Lherzolite Ol (57)thornOpx (25)thornCpx (15)thornSp (3) Porphyroclastic texture with Opx as porphyroclast (grain size of 2ndash3mm) Other

minerals are mostly51mm

SW13 Lherzolite Ol (65)thornOpx (22)thornCpx (10)thornSp (3) Porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts with

grain-sizes of 3ndash4mm

SW14 Lherzolite Ol (62)thornOpx (21)thornCpx (14)thornSp (3) Granuloblastic texture with grain-size of 51mm Serpentinization is developed

along fractures Minor basalt glass droplet is identified

SW15 Lherzolite Ol (62)thornOpx (24)thornCpx (13)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Fresh Ol Opx and Cpx form triple

junctions

SW16 Lherzolite Ol (66)thornOpx (23)thornCpx (10)thornSp (2) Granuloblastic texture with grain-size of 1ndash2mm Some Cpx develop reaction rims

weak serpentinization occurs along fractures

SW17 Lherzolite Ol (58)thornOpx (26)thornCpx (13)thornSp (3) Mosaic porphyroclastic texture Opx and sometimes Ol occur as porphyroclasts

with a grain-size of 3mm Most Cpx show resorption textures and Sp usually

coexists with the sieve-textured Cpx

SW18 Lherzolite Ol (72)thornOpx (16)thornCpx (10)thornSp (2) Mosaic porphyroclastic texture Opx and Ol occur as porphyroclasts with grain-size

of 3mm The relatively fine-grained minerals have grain-size of 1mm

SW19 Lherzolite Ol (60)thornOpx (16)thornCpx (22)thornSp (1) Granuloblastic texture with grain-size of 1mm Serpentinization is developed

along fractures in which some thin carbonate veins with width of 501mm

are identified

SW20 Lherzolite Ol (73)thornOpx (16)thornCpx (9)thornSp (2) Mosaic porphyroclastic texture with Opx occurring as porphyroclasts with grain-

size of 4mm

SW21 Wehrlite Ol (64)thornOpx (3)thornCpx (30)thornSp (4) Coarse and poikilitic texture The coarse-grained Ol has grain-size of 4ndash6mm

some fine-grained Ol occurs as inclusions within Opx Most Cpx show sieve

texture Serpentinite occurs along fractures with minor carbonate

SW22 Lherzolite Ol (61)thornOpx (26)thornCpx (11)thornSp (2) Granuloblastic texture with grain-size of 52mm The fresh Cpx rarely develops

sieve texture

Ol olivine Opx orthopyroxene Cpx clinopyroxene Sp spinel Numbers in parentheses are modal abundances calcu-lated from mass balance (Hermann amp Berry 2002)

CHU et al EASTERN NCC MANTLE THINING

1895

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 40: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

APPENDIX B ANALYT ICALMETHODSXRFApproximately 05 g of rock powder was mixed with36 g of Li2B4O7 and 3^4 drops of NH4Br and the mix-ture was fused in a furnace to form a glass disk The diskswere then analyzed using a Shimadzu XRF 1500 with acurrent and voltage of 50mA and 50 kV respectively Theanalytical results for Chinese ultramafic rock standardDZE-1 are reported inTable B1

ICP-MSAbout 100mg of crushed whole-rock powder were dis-solved in 2ml HF and 1ml HNO3 in a closed Teflonbomb which was heated on a hot plate at 1408C thenopened and subsequently evaporated to dryness to removesilica Then 15ml HF 15ml HNO3 and 10 ml HClO4

were added to theTeflon bomb The bomb was placed in asteel jacket and heated in an oven at 1908C for 5 days tocompletely dissolve the sample This was followed by dry-down and addition of concentrated HNO3 to form nitratesfollowed by a final evaporation Then 2ml 50 HNO3

was added to the Teflon bomb which was subsequentlyput it into the jacket again and heated to 1508C for 24 hFinally 05 g Rh (1ppm) was added to the solution as aninternal standard and then the solution was diluted by afactor of 500 The solutions were then ready for ICP-MSanalysis The analytical results for peridotite referencematerials JP-1 andWPR-1 are reported inTable B2

S analysisSamples were weighed into crucibles with Fe and W chipsadded as accelerators to ensure complete combustion ofthe sample The crucible was then heated in an RF induc-tion furnace in a stream of O2 and the sulfur derived asgaseous SO2 and SO3The gases were then passed througha dust filter drying tube (SO2 dissolves in water thuswater must be removed) and finally through a catalysttube to ensure conversion to dioxides From there thegases were directed into a sulfur IR absorption cell andthe SO2 peak was converted to S values These valueswere then adjusted for calibration and sample weight togive the final concentration

Sr^Nd^Hf isotopic analysisAbout 150^300mg of cpx powder was weighed into 7mlSavillexTM Teflon beakers and appropriate amounts ofmixed 87Rb^84Sr 149Sm^150Nd 176Lu and 180Hf spikeswere added The samples were dissolved using a mixedacid of 2ml HFand 02ml HClO4 on a hotplate at 1208Cfor more than 1 week After the samples were completelydissolved the solutions were dried on hotplate at 130^1808C to remove the HFand HClO4 The sample residues

were re-dissolved in 4ml of 6M HCl and then drieddown again Finally the samples were dissolved in 5ml of3M HClThe solutions were loaded onto pre-conditioned

Eichrom LN (LN-C-50A 100^150 mm 2ml) chromato-graphic columns The Rb Sr Sm Nd and matrix ele-ments were eluted with 3M HCl Then the Lu wasstripped with 4M HCl After this step Ti was removedwith a mixed solution of 4M HCl and 05 H2O2 ThenHf (and Zr) was stripped with a mixed acid of 6M HCland 02M HFThe Sr Nd collections from the LN columns were dried

down and re-dissolved in 1ml 25M HCl and then thesolutions were loaded onto pre-conditioned cationexchange columns packed with AG50W12 resins (200^400 mesh 2ml) Rubidium and Sr were stripped with5M HCl and then REE were stripped with 6M HClThe REE collections were dried down and re-dissolved

in 04ml 025M HCl and then the solutions were loadedonto pre-conditioned Eichrom LN (LN-C-50B 100^150 mm 2ml) chromatographic columns The La Ce anda portion of the Pr were eluted with 6ml 025M HCl andthe Nd was stripped with 3ml 04M HCl and then theSm was stripped with 4ml 08M HCl

Re^Os and PGE analysisAbout 2 g of finely ground sample powder (200^400 mesh)and appropriate amounts of a 187Re^190Os mixed spikeand a 191Ir^99Ru^194Pt^105Pd mixed spike were weighedinto a clean dry chilled Pyrex borosilicate glass Cariustube Then 3ml of purified concentrated HCl and 6ml ofpurified concentrated HNO3 were added into the tubeand then the tube was sealed (Shirey amp Walker 1995) Thedigestions were performed at 2408C for 48^72 h in anoven at IGGCAS and at 2608C for 48^72 h in an ovenat UMD After opening the tubes Os was extracted fromthe aqua regia solution into CCl4 (Cohen amp Waters 1996)and then back-extracted into HBr followed by purificationvia microdistillation (Birck et al 1997) The total proce-dural Os blank was 3^5 pg with a 187Os188Os of about015 at IGGCAS and about 05 pg with a 187Os188Os ofabout 017 at UMD respectively The blank correction onOs concentrations and Os isotopic ratios was negligiblefor all samplesRhenium and remaining PGE (Ir Ru Pt Pd) were

separated from the matrix and purified by anion exchangechromatography using 2ml resin (AG 18 100^200mesh) (Walker et al 2008) The Re Ru fraction waseluted with 6M HNO3 Ir and Pt were eluted with 135MHNO3 and the Pd was eluted with 10M HCl The Re wasfurther purified using a small anion exchange columnpacked with 01ml resin at UMD All the collected frac-tions were dried down and re-dissolved in 1ml of 08MHNO3 for ICP^MS measurement Total procedural

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1896

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 41: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

blanks were about 3 pg for Re 7 pg for Ir 7 pg for Ru 4 pgfor Pt and 4 pg for Pd at IGGCAS and about 5 pg for Re4 pg for Pd 10 pg for Pt 1pg for Ir and 3 pg for Ru at

UMD The blank corrections were negligible (51) forIr Ru Pt and Pd but were as great as 10^70 for Re forsome low-Re samples

Table B1 Analytical results of Chinese ultramafic rock standard DZE-1 by XRF (wt )

SiO2 TiO2 Al2O3 TFe2O3 MnO MgO CaO Na2O K2O P2O5 LOI Total

Measured 3454 001 075 688 007 4095 003 001 004 001 1520 9849

Reference value 3434 0008 067 69 0068 4103 01 0008 001 0004

Source Govindaraju (1994)

Table B2 Analytical results of ultramafic rock standard materials by ICP-MS (ppm)

Element JP-1 WPR-1

Measured References Measured References

(nfrac14 1) Dulski (2001) Qi et al (2005) Makishima et al (2006) GeoREM (nfrac14 1) GeoREM

Ni 2508 2460 2780 2900

Cu 50 57ndash672 1514 1640

Zn 50 43ndash61 918 95

Sr 0605 063 0546 705 7

Y 0088 0097 0088 0082 418 5

Zr 56 55 192 18

Nb 0042 00336ndash1 197 24

Ba 905 92 95 256 22

La 0029 0034 0031 0027 203 22

Ce 0060 0063 0054 0060 563 6

Pr 00078 00089 00076 00074 0773 07

Nd 0033 0033 0030 00318 359 35

Sm 00097 0009 00077 00084 0907 09

Eu 00029 00021 00011 00024 0294 031

Gd 0012 00092 00064 00097 100 09

Tb 00024 00016 00017 00019 0155 01ndash02

Dy 0015 0013 0013 0015 0938 11

Ho 00038 0003 00031 00036 0184 018

Er 0012 0011 0011 0012 0511 05

Tm 00024 00023 00021 00024 00711 009

Yb 0022 0021 0019 0021 0466 048

Lu 00046 0004 00034 00042 00716 007

Hf 013 0127 0579 061

Ta 00035 000304ndash0068 0117 021

Pb 0128 018 00765 525 6

Th 0010 0012 0013 0329 04

U 0014 00116 0014 0182 017

GeoREM httpgeoremmpch-mainzgwdgde

CHU et al EASTERN NCC MANTLE THINING

1897

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898

Page 42: Temporal Evolution of the Lithospheric Mantle beneath the ... · mation, mineralization, and igneous activity occurred, including extensive eruption of intermediate^acid volcanic

Fig B1 Comparison of analytical results of Os isotopic ratios between UMD and IGGCAS UMD measured at University of MarylandIGGCAS measured at Institute of Geology and Geophysics Chinese Academy of Sciences

JOURNAL OF PETROLOGY VOLUME 50 NUMBER 10 OCTOBER 2009

1898