Spectral properties and mineral compositions of acapulcoite–lodranite clan

meteorites: Establishing S-type asteroid–meteorite connections

Michael P. LUCAS 1*, Joshua P. EMERY1, Takahiro HIROI2, and Harry Y. MCSWEEN 1

1Department of Earth & Planetary Sciences, University of Tennessee, 1621 Cumberland Ave., 602 Strong Hall, Knoxville,

Tennessee 37996, USA2Department of Earth, Environmental and Planetary Sciences, Brown University, Providence, Rhode Island 02912, USA

*Corresponding author. E-mail: mlucas9@vols.utk.edu

(Received 23 December 2017; revision accepted 07 August 2018)

Abstract–Except for asteroid sample return missions, measurements of the spectralproperties of both meteorites and asteroids offer the best possibility of linking meteoritegroups with their parent asteroid(s). Visible plus near-infrared spectra reveal distinguishingabsorption features controlled mainly by the Fe2+ contents and modal abundances ofolivine and pyroxene. Meteorite samples provide relationships between spectra andmineralogy. These relationships are useful for estimating the olivine and pyroxenemineralogy of stony (S-type) asteroid surfaces. Using a suite of 10 samples of theacapulcoite–lodranite clan (ALC), we have developed new correlations between spectralparameters and mafic mineral compositions for partially melted asteroids. A well-definedrelationship exists between Band II center and ferrosilite (Fs) content of orthopyroxene.Furthermore, because Fs in orthopyroxene and fayalite (Fa) content in olivine are wellcorrelated in these meteorites, the derived Fs content can be used to estimate Fa of thecoexisting olivine. We derive new equations for determining the mafic silicate compositionsof partially melted S-type asteroid parent bodies. Stony meteorite spectra have previouslybeen used to delineate meteorite analog spectral zones in Band I versus band area ratio(BAR) parameter space for the establishment of asteroid–meteorite connections with S-typeasteroids. However, the spectral parameters of the partially melted ALC overlap with thoseof ordinary (H) chondrites in this parameter space. We find that Band I versus Band IIcenter parameter space reveals a clear distinction between the ALC and the H chondrites.This work allows the distinction of S-type asteroids as nebular (ordinary chondrites) orgeologically processed (primitive achondrites).

INTRODUCTION

Significant progress has been made during the lastseveral decades in linking stony (S-complex) asteroids totheir olivine- and/or pyroxene-rich stony meteoritecounterparts. Initial work to formulate correlationsbetween the spectral parameters of mafic silicateassemblages and their mineralogical properties utilizedlaboratory visible (VIS) plus near-infrared (NIR)reflectance spectra of natural terrestrial olivine andpyroxene mixtures (Adams 1974; Cloutis et al. 1986;King and Ridley 1987). More recent studies make useof mineralogical and spectral properties of meteorite

samples to develop equations useful for estimating theolivine and pyroxene mineralogy of asteroid surfacesfrom the band parameter analysis of their VIS+NIRspectra. In particular, these studies have demonstratedthat Fe2+ compositions of olivine Fa (mol% fayalite;Fe2SiO4), and pyroxene Fs (mol% ferrosilite; FeSiO3),and modal mineral abundances of these minerals can beextracted from the reflectance spectra of someS-complex asteroids. Mineralogic equations based onstony meteorites for the estimation of S-complexasteroid surface mineralogy have been published forordinary chondrites (H, L, LL) (Burbine et al. 2003;Dunn et al. 2010a); howardite, eucrite, diogenite (HED)

Meteoritics & Planetary Science 54, Nr 1, 157–180 (2019)

doi: 10.1111/maps.13203

157 © The Meteoritical Society, 2018.

meteorites (Burbine et al. 2007, 2009; Moskovitz et al.2010); and olivine-dominated (R chondrite, brachinite,pallasite) meteorites (Reddy et al. 2011; Sanchez et al.2014). These equations help to correlate S-complexasteroids to meteorite types that possess diversepetrologic histories: unmelted nebular materialrepresented by chondrites; and magmatic (totallymelted) achondrites represented by HED meteorites.The petrologic histories of rare olivine-dominatedasteroids are not as evident. They may represent:pallasites (mixtures of olivine cumulates and FeNimetal, which are thought to represent asteroidal core–mantle boundaries; Sunshine et al. 2007), R chondrites(olivine-rich nebular material; Sanchez et al. 2014),brachinites (olivine-rich partial melt residues; Sunshineet al. 2007; Sanchez et al. 2014), or a combination ofthe above. In the near future, only a few bodies will bevisited for asteroid sample return. In the meantime,spectral and mineralogical studies of meteorites make itpossible to derive S-complex asteroid surfacemineralogies using solely ground-based observations.

Gaffey et al. (1993) categorized the spectralparameters of S-complex asteroids into sevenmineralogical subtypes that generally cluster near theolivine–orthopyroxene mixing line of Cloutis et al.(1986): olivine-dominated objects (S[I]), olivine–pyroxene mixtures representing ordinary chondrites andprimitive achondrites (S[II] to S[VI]), to nearly purepyroxene or pyroxene–feldspar–metal mixtures (S[VII])(Fig. 1). These classifications are possible because S-complex asteroids reflect near-infrared photons thatcarry strong ~1 lm (Band I) and/or ~2 lm (Band II)absorptions caused by the presence of Fe2+ in olivineand pyroxene (Burns 1993). Band I center is sensitive tothe iron contents of olivine and pyroxene and theirmodal abundances, and the ratio of the two band areas(BAR; Band II area/Band I area) is sensitive to theirmodal abundances (Cloutis et al. 1986). Gaffey et al.(1993) proposed potential meteorite analogs for each ofthe seven subtypes plus a basaltic achondrite (BA) zone(Fig. 1). The BA zone is associated with howardite–eucrite–diogenite (HED) meteorites that are generallyaccepted as originating from 4 Vesta and its collisionalfamily (Vestoids) (McCord et al. 1970; Gaffey 1997;McSween et al. 2013; Russell et al. 2015). The nebularordinary chondrite (OC) class of meteorites is associatedwith the S(IV) mineralogical zone (“OC boot”) (Fig. 1)(Gaffey et al. 1993, 2002). Significant progress has beenmade in extracting detailed mineralogic interpretationsfor S-complex asteroids that possess spectral propertiesconnected with the S(IV) zone. Burbine et al. (2003)derived an equation for determining the olivine andpyroxene modal abundances of OCs from analysis ofBAR. This correlation was modified by Dunn et al.

(2010a, 2010b) who added equations for determiningthe compositions of olivine (Fa) and pyroxene (Fs) inOCs from analysis of Band I centers. These equationssuccessfully discriminate the H, L, and LL groups ofthe OC class. Despite these improvements in formingasteroid–meteorite links for nebular (OCs) and igneous(BA) material, associations between S-complex asteroidsand partially melted (primitive achondrite) meteoriteshave not been established. Primitive achondrites possesssilicate mineralogies composed predominantly of olivineand pyroxene and have been associated with some S-complex asteroids (Hiroi et al. 1993; Gaffey et al. 1993;McCoy et al. 2000; Burbine et al. 2001; Dunn et al.2013; Lucas et al. 2016). Yet, because there have notbeen many laboratory spectral measurements ofprimitive achondrites meteorites, the association remainsvague.

Primitive achondrites include several groups ofstony meteorites that have mineral and bulkcompositions that are similar to chondrites in manyaspects, but have achondritic textures formed asresidues from low degrees of partial melting (Prinz et al.1983). These groups include the ureilites, brachinites,winonaites, acapulcoites, and lodranites. The ureilitesare thought to be residues from the partial melting/smelting of carbonaceous chondrite-like precursormaterial (Warren and Kallemeyn 1992; Singletary andGrove 2006). Ureilites have experienced at least 15%partial melting (Warren and Kallemeyn 1992; Krotet al. 2007), generally higher than that of theacapulcoite and lodranite groups. Brachinites areolivine-rich (~80–90 vol%) primitive achondrites(Mittlefehldt et al. 2003). Early models of origin forbrachinites include the metamorphism and oxidation ofchondritic material (Nehru et al. 1996), though furtherstudy of more recent brachinites has concluded that

Gaffey et al. (1993) Asteroid S-subtypes“olivine-dominated”

“OC boot”

“basaltic achondrite” (BA)

S(II) through S(VI) “ol+px mixtures”

Fig. 1. Asteroid S-subtypes (S[I] through S[VII]) and thebasaltic achondrite (BA) spectral regions of Gaffey et al. (1993)shown in Band I center (~1 lm) versus band area ratio (BAR)parameter space; the thick black curve indicates the location ofthe olivine–orthopyroxene mixing line of Cloutis et al. (1986).(Color figure can be viewed at wileyonlinelibrary.com.)

158 M. P. Lucas et al.

their petrology and geochemical characteristics indicatean igneous origin (Mittlefehldt et al. 2003). Thewinonaites are a small group (~30 meteorites) ofprimitive achondrites that exhibit chondriticmineralogies and bulk compositions. They have oxygenisotopic compositions similar with two groups ofsilicate-bearing iron meteorites (IAB and IIICD),suggesting the formation of these meteorites on acommon parent body (i.e., a clan) (Benedix et al. 1998;Krot et al. 2007). Primitive achondrites of theacapulcoite–lodranite clan (hereafter ALC) have beendescribed by McCoy et al. (1997b, 2000), andMittlefehldt et al. (1996). Acapulcoites are essentiallychondritic in composition, but have experienced verylow degrees of partial melting (<1 to ~5%). Thesemeteorites have been heated to ~950–1050 °C, warmenough to induce FeNi-FeS cotectic melting (McCoyet al. 1997b). They exhibit granular, recrystallizedtextures, and are fine-grained with grain intersticestypically filled with FeNi metal, indicative of thisheating. Lodranites have larger grain-sizes and showhigher degrees of partial melting (~5 to 20%) than theacapulcoites, enough for melt pockets to growsubstantially to form interconnected channels,conducive for melt segregation. These meteoritesexperienced temperatures of ~1050–1200 °C, hot enoughto produce basaltic partial melts (McCoy et al. 1997a).Accordingly, they display equigranular textures withnumerous 120° triple junctions (McCoy et al. 2000).Oxygen isotope (McCoy et al. 1996; Clayton et al.1992) and cosmic ray exposure age (CRE) data (Eugsterand Lorenzetti 2005; Patzer et al. 2003) for acapulcoitesand lodranites overlap and point to a single parentbody for the clan (Mittlefehldt et al. 1996; McCoy et al.1997a). These meteorites are posited to representsamples from a lithologically diverse, partiallydifferentiated asteroid.

Several models have been developed to explain thepetrologic and chemical properties of these partiallymelted rocks. The ALC display properties of acontinuous range of whole-rock partial melting (<1 to~20%) that reflect parent-body scale melting on theacapulcoite–lodranite parent body (ALPB). Yet,petrologic and geochemical evidence are consistent withmelting of heterogeneous precursor materials and morelocalized igneous processes (Mittlefehldt et al. 1996;McCoy et al. 2000). Mittlefehldt et al. (1996) arguedthat igneous heating on the ALPB was heterogeneous inboth space and time, and that melting on a parent-bodyscale (e.g., resembling Vesta and the HEDs) probablydid not occur. McCoy et al. (1997a) concluded thatacapulcoites and lodranites represent the residues fromdifferent degrees of partial melting of common, yetisotopically and chemically heterogeneous chondritic

precursor materials. Lodranites experienced temperatureshigh enough to segregate basaltic partial melts(plagioclase–pyroxene). The lack of preservation of thesepartial melts or their crystallized products (basalts) maysuggest that they were removed from the asteroid parentbody early in solar system history. McCoy et al. (1997b)evoked pyroclastic volcanism to explain the missingbasalts that would be complementary to the residuallodranites. However, removal by the gradual orcatastrophic comminution of an ALPB basaltic crustcannot be ruled out. Eugster and Lorenzetti (2005)proposed a two-layer “onion shell” model that suggestsacapulcoites formed from an earlier and faster coolingouter layer, while lodranites formed within a slower andlater cooling inner layer of the ALPB. Rubin (2007)posited that the petrogenesis of the ALC clan involvedshock melting of a CR-like carbonaceous chondriteprecursor, with subsequent postshock annealing takingplace beneath an insulating regolith to explain the lowshock-stage rocks of the clan.

Trace element (Floss 2000a) and bulk-rock (Patzeret al. 2004) chemical analyses show an entire range ofdifferent metamorphic degrees, from primitiveacapulcoites that have textures similar to chondrites(some with relict chondrules) and near chondriticcompositions, to lodranites that have recrystallizedtextures and experienced segregation of silicate melts.These analyses show that the petrologic diversity of theALC is more complex than simple division into eachmeteorite group. Consequently, Patzer et al. (2004)subdivided the ALC into five petrologic types based onincreasing temperature and migration of eutectic metal-sulfide and silicate melts, features that help to explaintheir varying degrees of petrologic evolution. They are(1) primitive acapulcoites, (2) typical acapulcoites, (3)transitional acapulcoites, (4) enriched acapulcoites, and(5) lodranites. Small-scale melt migration processeswithin these rocks would have contributed to themineralogical heterogeneity of the ALPB (Patzer et al.2004).

Multiple studies have demonstrated that the ALPBshould be spectrally related to some asteroid subtype ofthe S-complex (Gaffey et al. 1993; Hiroi et al. 1993;McCoy et al. 1997b, 2000; Burbine et al. 2001). Beforethe Near-Earth Asteroid Rendezvous spacecraft (NEARShoemaker) analyzed near-Earth asteroid (NEA) 433Eros (Cheng 2002), there was speculation that 433 Erosmight have experienced some degree of meltinganalogous to primitive achondrites (McCoy et al. 2000;Burbine et al. 2001). However, mission results indicatedhomogeneous spectral properties across its surfaceconsistent with an ordinary chondrite composition(McCoy et al. 2001; Cheng 2002). One clue regardingthe search for the source of the ALPB is that ALC

Spectra and mineralogy of acapulcoites and lodranites 159

meteorites possess short CRE ages (~4–7 Myr) anddisplay a tight cluster of values (Eugster et al. 2006).These values indicate an “express” delivery sourceregion for the clan. Dunn et al. (2013) found a probablesource region at the Jovian 3:1 mean-motion resonancefor one primitive achondrite-like NEA. Interestingly, Hchondrites are thought to be sourced from the 3:1mean-motion resonance (Gaffey and Gilbert 1998;Thomas and Binzel 2010), and this group also has shortCRE ages (median ~7 Myr; Marti and Graf 1992) thatoverlap that of acapulcoite–lodranites (Eugster andLorenzetti 2005). Despite the similarities in CRE agesbetween the ALC and H chondrites, the source region(s)for the ALPB remains unclear.

Although a candidate for the ALPB has not beenidentified, the spectral properties of some S-complexasteroid types have olivine+pyroxene mineralogiesconsistent with the clan. The chemical and mineralogicproperties of ALC meteorites are well known. However,the difficulty in the identification of plausible parentasteroids for the ALC is in part due to the lack of high-quality reflectance spectra of these uncommonmeteorites. To address this shortcoming, we acquiredhigh-quality VIS+NIR reflectance spectra of 10 ALCmeteorites after the neutralization and removal ofterrestrial weathering products. We performed detailedspectral band parameter measurements of these spectra.In order to derive more robust relationships betweenthe spectral parameters and mineral compositions ofacapulcoites–lodranites, we also measured the mineralcompositions of their silicate phases using electronprobe microanalysis (EPMA). The chemical andspectral analyses of acapulcoite–lodranite meteoritespresented here provide additional data to help identifyasteroids that have mineralogies consistent with theALC. Our comparison of the spectral andcompositional features of these samples yield spectralcalibrations to provide more accurate mineralogicinterpretations for S-complex asteroid spectra, whichmay make it possible to spectrally distinguish otherpartially differentiated asteroids.

ACAPULCOITE–LODRANITE SAMPLES

A suite of 10 meteorites of the acapulcoite–lodraniteclan were obtained for this study. The samples werethen prepared for major element analysis of mineralsand for the acquisition of VIS+NIR reflectance spectra.Many meteorites of the ALC have been wellcharacterized geochemically; however, there is a lack ofdata regarding VIS+NIR reflectance spectra for thesepartially melted rocks. All of the samples analyzed inthis study are classified as meteorite “finds”; theeponymous meteorites of the ALC, Acapulco, and

Lodran, respectively, are the only known “falls” amongthe clan (Meteoritical Bulletin Database; https://www.lpi.usra.edu/meteor/, accessed June 4, 2017). Oursample suite includes four acapulcoites, five lodranites,and one transitional acapulcoite–lodranite. Transitionalmeteorite LEW 86220 is an acapulcoite in compositionand grain size, and also contains a discrete coarse-grained gabbroic lithology (McCoy et al. 1997b; Floss2000b).

Four Antarctic samples were provided by theNASA Meteorite Working Group (MWG) (Table 1).We acquired six meteorite samples commercially, fivefound in Northwest Africa (NWA), and one samplefound in Dhofar, Oman (Table 1). For the Antarcticsamples, we obtained both bulk chip specimens(~500 mg) and polished thin sections. For the sixcommercial samples, we obtained bulk unpolished slabs(~1–2 g) in order to have enough material forpowdering and polished thick section preparation.Sample characteristics are listed in Table 1.

Petrographic Sample Preparation

Polished thin or thick sections were prepared forpetrographic and major element analyses of mineralsfor the 10 acapulcoite–lodranite samples. The Antarcticspecimens were prepared as polished thin sections of~30 lm thickness mounted on 1-inch acrylic disks. Thecommercially acquired specimens were cut into thicksections using a low-speed petrographic saw withcutting oil as a lubricant in order to avoid oxidationof the specimens. One surface of each thick sectionwas chosen as the analytical surface, and the sectionwas cast in acrylic for electron probe microanalysis.The analytical surface of each specimen was firstground using aluminum oxide sanding disks to 400 grit(~22 lm) size, then polished to a final surface of~1 lm using diamond polishing paste of threeprogressively smaller particle sizes. Water was not usedin any part of the process. Rather, oil and reagentalcohol were used for the cutting, grinding, andpolishing stages.

Reflectance Spectroscopy Sample Preparation

We prepared meteorite powders (<125 lm) from thebulk chips or unpolished slabs of each sample for use inthe acquisition of VIS+NIR reflectance spectra. Samplepreparation techniques were consistent for all specimensand we thoroughly cleaned all instruments withisopropanol (99.9% assay) after handling each specimento prevent cross-contamination. These meteoritescontain from <1 to 29 vol% metal. Metal content canalter the slope of reflectance spectra and suppress

160 M. P. Lucas et al.

absorption band depth (Cloutis et al. 2015). However,the band parameters used to derive the mafic silicatemineralogy of our ALC meteorites for comparison to S-type asteroids (band centers and BAR) are largelyinsensitive to metal contents (Cloutis et al. 2010, 2015).Analysis of these band parameters includes division ofthe absorption features by a straight-line continuum;therefore, band parameters such as band centers andBARs are insensitive to continuum slope variationsproduced from metal content (Cloutis et al. 1986;Gaffey 2010). Cloutis et al. (2015) showed that Band Iposition is principally unaffected by metal abundancesof <80 wt%. Furthermore, although increasing metalcontent can suppress band depth, thereby reducing bandarea, using the ratio of Band II area/Band I area (BAR)effectively cancels out any reduction in band area dueto metal content. We therefore removed the metallicfraction of each acapulcoite–lodranite sample duringour powdering procedure, in order to acquire reliablespectra of the silicate fraction.

The meteorites were coarsely crushed (~250–500 lm) and placed in a glass culture dish withisopropanol. The metal fraction of each meteorite wasisolated from the silicate fraction by passing a strongmagnet underneath the culture dish. The metal fractionwas then removed from the culture dish with tweezers,leaving behind the silicate fraction and weatheredmaterial. This method was performed serially (3–59) toremove a maximum amount of metal from each sample.After drying, masses of the separated metal and silicatefractions were measured. We typically recovered90–95% of the initial mass of each bulk specimen usedfor powdering.

In order to compare the reflectance spectra ofmeteorites to asteroids, we powdered the remainingsilicate fractions in an agate mortar and pestle and drysieved the powder to <125 lm (Burbine et al. 2003).Asteroid surfaces are generally blanketed in regolith,therefore meteorite powders of <125 lm provide asuitable analogue in the laboratory. For a detailedtreatment regarding meteorite sampling and preparation,see the Appendix in Jarosewich (1990).

Ethanolamine Thioglycolate Treatment

The spectral signatures of meteorite “finds” areaffected by terrestrial weathering during the longintervals spent in Antarctic ice or on the desert floor(Hiroi et al. 1993). For meteorites that containconsiderable amounts of metallic FeNi, weathering istypically expressed through the oxidation of metal intoiron hydroxides (e.g., limonite). All 10 acapulcoite–lodranite samples suffer from variable amounts ofterrestrial alteration (Table 1).

Our method of neutralization and removal of theseweathering products used an ethanolamine thioglycolate(EATG) solution (Cornish and Doyle 1984) that, to ourknowledge, has not been previously applied toreflectance spectroscopy of meteorite powders. TheEATG leaching method has been used to remove ironhydroxide alteration products from weatheredmeteorites for oxygen isotope studies (Martins et al.2007; Greenwood et al. 2012). This method is preferredto conventional leaching using dilute HCl, as treatmentin acid can partially remove feldspathic silicates(Greenwood et al. 2012). Experience by one of us

Table 1. The sample suite of acapulcoite–lodranite meteorites analyzed in this study.

Meteorite nameaMeteoritegroup

Weathergradeb

Shockstage

Originalmass (g)

Bulk samplemass Sample typec RELAB ID

Acapulcoite

MET 01212,14 Acapulcoite B/C – 31.37 0.522 PM,CP MT-JPE-301-ADhofar 125 Acapulcoite W1/2 S1 2697 0.916 UPS MT-JPE-304-ANWA 725 Acapulcoite – – 3820 2.435 UPS MT-JPE-305-A

NWA 2871 Acapulcoite W3 Low 3467 1.498 UPS MT-JPE-306-ALodraniteLAR 06605,8 Lodranite B – 34.61 0.488 PM,CP MT-JPE-300-A

MAC 88177,53 Lodranite B/C S5 35.27 0.393 PM,CP MT-JPE-302-ANWA 5488 Lodranite – – 110 1.588 UPS MT-JPE-308-ANWA 7312 Lodranite Low Low 778 2.145 UPS MT-JPE-309-ANWA 7674 Lodranite Low Low 225 2.266 UPS MT-JPE-310-A

TransitionalLEW 86220 Acap/Lod – – 25.04 0.436 PM,CP MT-JPE-314-A

aMET = Meteorite Hills; NWA = Northwest Africa; LAR = Larkman Nunatak; MAC = MacAlpine Hills; MIL = Miller Range; LEW = Lewis

Cliff.bLetter (A–D) weathering grade for Antarctic hand specimens; numerical (W) weathering grade for polished sections from Meteoritical Bulletin.cUPS = unpolished slice; PM = polished thin section; CP = chip.

Spectra and mineralogy of acapulcoites and lodranites 161

(T. Hiroi) has shown that treatment with dilute HCl canalso result in a residual H2O absorption band near2 lm that is difficult to remove for particulate samples.

We applied the EATG leaching method to thepowders (<125 lm) of all samples and acquired theirVIS+NIR reflectance spectra both before and aftertreatment. The effectiveness of this method in removingiron hydroxide alteration from meteorite powders canbe observed qualitatively from photographs of thepowder specimens (Fig. 2). Individual samples wereplaced in glass scintillation vials and leached in asolution of 15% EATG in isopropanol (99.9% assay)for a minimum of 4 h. The thioglycolate anion willreact with Fe3+ in solution to form a violet-coloredferrothioglycolate complex (Cornish and Doyle 1984).The intensity of this colored-complex serves as a handy

gauge of the degree of weathering for each specimen.The more weathered samples were given a secondtreatment in 5% EATG solution for a minimum of 2 h.The samples were washed several times withisopropanol and then dried for spectral analysis.

To investigate possible effects that the EATGleaching method may have on spectral band parameters,we applied the same EATG method on sample powders(<125 lm) of natural olivine and pyroxene controlsamples. We measured the spectral band parameters forthese controls both before and after treatment withEATG. Band parameter results for these controlsamples are discussed in the EATG-Treated Olivine andOrthopyroxene Control Samples section below.Additionally, to compare our EATG “cleaned” meteoritespectra with spectra of “fresh” acapulcoite–lodranite

Fig. 2. Lodranite LAR 06605 (left) and acapulcoite MET 01212 (right) meteorite powders (<125 lm) shown within sampledishes before and after removal of terrestrial weathering products. RELAB IDs for each sample shown under sample dish. a)Above—sample powders before treatment with EATG show significant iron hydroxide staining. b) Below—the same samplepowders after treatment with EATG to remove terrestrial alteration. Iron hydroxide staining is considerably reduced with EATGtreatment.

162 M. P. Lucas et al.

falls, we have included in our analysis the reflectancespectra of the Acapulco and Lodran meteorites from theRELAB spectral database (http://www.planetary.brown.edu/relabdata/).

EATG-Treated Olivine and Orthopyroxene Control

Samples

In order to demonstrate that measured band centersof the mafic silicates found within our samples are notaffected by the EATG leaching procedure, we applied theEATG method to powders (<125 lm) of natural mineralspecimens that we used as control samples. The formationof a faint violet-colored ferrothioglycolate complexduring the leaching procedure verified that the controlsamples contained a minor amount of oxidized material.

We acquired the VIS+NIR reflectance spectra (seethe VIS+NIR Bidirectional Reflectance Spectroscopysection) of the following natural mineral specimens bothbefore and after treatment with EATG: San Carlos

olivine (Fa10), an orthopyroxene (enstatite) sample fromTanzania (Fs9), and a 50/50 wt% mixture of the twominerals. Untreated versus EATG-treated sample spectraare overplotted (reflectances not offset) for each of thethree controls in Fig. 3. The EATG-treated samplesshow a slight increase in band depth for both Band I andBand II. This small increase may be the result of theremoval of a small amount of weathering productsduring the leaching procedure. The band centers areessentially unaffected by the EATG treatment.

To quantify these observations, we measured theband parameters of both untreated and treated controlsamples using the same methods (see the Spectral BandParameter Measurements section) used for theacapulcoite–lodranite meteorites. The results are givenin Table 2. EATG-treated samples show increased banddepths and therefore an increase in band area. Similarto the results observed for the meteorites (see theAcapulcoite–Lodranite Spectral Results section),measured Band I centers for EATG-treated controlsamples are identical to the untreated values. MeasuredBand II centers for EATG-treated controls only showsmall variation (≤0.01 lm) from untreated values. Weconclude that band center positions of olivine andorthopyroxene are unaffected by the application of theEATG weathering removal method.

ANALYTICAL METHODS

Petrography

Petrographic analysis of four polished thin sectionsand six polished thick sections (mounted on 1 inch acrylicrounds) was performed using a polarizing light microscope.After noting petrographic descriptions, we prepared high-resolution photomosaic images of each sample section. Forthe thin sections, we created three photomosaics, one eachin plane-polarized (PPL), cross-polarized (XPL), andreflected light (RL). For the thick sections, we created onlyRL photomosaics. We used low power (25x) under aplanachromat objective with a Nikon DS-Fi1 camera torecord high-resolution photomicrographs. These imageswere then stitched together in software (NIS Elements2.30) to construct the photomosaics. XPL photomosaics ofacapulcoite MET 01212 and transitional acapulcoite LEW86220 are shown in Fig. 4. Relevant petrologic descriptionsand modal mineral abundances for acapulcoites andlodranites from our analysis and the literature are insupporting information.

Electron Probe Microanalysis

We measured major elements in minerals withelectron probe microanalysis (EPMA), using the

0.5 1.0 1.5 2.0 2.5Wavelength (µm)

0.0

0.5

1.0

1.5

2.0

2.5N

orm

aliz

ed R

efle

ctan

ce

San Carlos Olivine

Tanzania Enstatite

Ol+En 50/50 mixture (wt%)

Untreated

EATG Treated

Fig. 3. VIS+NIR reflectance spectra of olivine and enstatiteminerals and a 50/50 wt% mixture of these minerals used ascontrol samples for the EATG weathering removal procedure.Untreated sample spectra reflectance values are shown as solidblack circles, while EATG reflectance values are shown asopen colored triangles. Reflectance values of different samplesare offset by 0.75 reflectance for clarity; however, reflectancevalues for the same samples (untreated and EATG treated) arenot offset. (For interpretation of the references to color in thisfigure caption, the reader is referred to the web version of thisarticle.) (Color figure can be viewed at wileyonlinelibrary.com.)

Spectra and mineralogy of acapulcoites and lodranites 163

Cameca SX100 electron microprobe at the University ofTennessee. Quantitative analyses were carried out viawavelength dispersive spectrometry (WDS), using a15 kV accelerating voltage, beam currents of 10–30 nA,and a spot size of 1 lm. Standards, which include bothnatural (olivine, diopside, hematite, spessartine, spinel,and rutile) and synthetic (Cr and Ni metal) referencematerials, were measured regularly during analyticalsessions to ensure data quality. Background and peakcounting times for the major elements were typically20–30 s, and ZAF corrections were applied using theCameca PAP procedure. Mineral phases varied fromsample to sample, but when present, the followingmineral phases were analyzed using WDS mode: olivine(Ol), orthopyroxene (Opx), clinopyroxene (Cpx),plagioclase (Pl), spinel (Spl), and apatite (Ap). Wetypically did not analyze iron–nickel (FeNi) metal or ironsulfides (FeS, or troilite). For each meteorite, we analyzedthe cores of multiple grains (typically n = 10–12) of majormineral phases and at least three spots per grain forminor mineral phases. To check for mineral zoning, weanalyzed transects (typically 10–20 spots) acrossrepresentative grains for all major mineral phases in eachmeteorite sample. Detection limits for silicates andspinels were: <0.03 wt% for SiO2, MgO, TiO2, Al2O3,CaO, and Na2O; <0.04 wt% for K2O and Cr2O3; <0.05wt% for FeO, MnO, and P2O5; and <0.06 wt% forNiO. Electron backscatter (BSE) images were obtainedof representative phases for each meteorite specimen.Our EPMA results for major element mineralcompositions of acapulcoite–lodranite meteorites arepresented in Tables 3 through 6.

VIS+NIR Bidirectional Reflectance Spectroscopy

We measured reflectance spectra of the powderedsilicate fraction of each meteorite using the bidirectionalspectrometer (0.32–2.55 lm) at Brown University’sKeck/NASA Reflectance Experiment Laboratory(RELAB). For a detailed discussion of the capabilitiesof the instruments at RELAB, see Pieters and Hiroi(2004). The spectra were collected under standardoperating parameters, which include a spectralresolution set to 0.01 lm for each of the four gratingsthat cover the wavelength range, and a viewinggeometry of 30° incidence and 0° emergence angle. Wecollected reflectance spectra for all meteorite powdersboth before and after EATG treatment (see theAcapulcoite–Lodranite Spectral Results section). TheRELAB IDs of our sample suite are listed in Table 1.The RELAB sample IDs of the namesake “falls” of theacapulcoite and lodranite groups are: Acapulco(TB-TJM-043) and Lodran (TB-TJM-041).

Spectral Band Parameter Measurements

We measured the band parameters of the strong 1and 2 lm absorption bands (Band I and II,respectively) for eight meteorite spectra from oursample suite, and also the meteorites Acapulco andLodran. We used an Interactive Data Language (IDL)-based code, the Spectral Analysis Routine for Asteroidsor SARA (Lindsay et al. 2015) to measure Band I andII centers, depths, areas, band area ratios, and theiruncertainties. The band area ratio (BAR) is defined as

Table 2. Band parameter analysis for three EATG leaching method mineral control samples (ground particulate<125 lm fraction; RELAB ID suffix “A” for untreated samples, “T” for EATG-treated samples).

ControlSample Description RELAB ID BIC

BICerror

BI area(910�2)

BI areaerror(910�2) BIIC

BIICerror

BII area(910�2)

BII areaerror(910�2) BAR

BARerror

San Carlosolivine

Untreated OL-JPE-020-A 1.06 0.001 30.6 0.04 – – – – – –

San Carlosolivine

EATGtreated

OL-JPE-020-T 1.06 0.003 33.8 0.06 – – – – – –

Tanzaniaenstatite

Untreated PX-JPE-020-A 0.910 0.001 16.2 0.04 1.830 0.003 32.5 0.06 2.01 0.01

Tanzania

enstatite

EATG

treated

PX-JPE-020-T 0.910 0.001 18.5 0.04 1.820 0.002 38.4 0.06 2.08 0.01

Ol+En 50/50mixture

(wt%)

Untreated MX-JPE-019-A 0.910 0.001 17.9 0.04 1.840 0.001 14.8 0.05 0.825 0.003

Ol+En 50/50mixture

(wt%)

EATGtreated

MX-JPE-019-T 0.910 0.001 21.6 0.04 1.843 0.001 17.1 0.05 0.793 0.003

BIC = Band I center; BIIC = Band II center; BAR = Band Area Ratio.

164 M. P. Lucas et al.

the ratio of (Band II area)/(Band I area). The SARAalgorithm computes linear continua across the broadBand I and II absorptions by fitting 5th-orderpolynomials to the absorption band shoulders and thenfitting linear continua tangentially to the shoulders, andthen divides the spectral segments by their respectivecontinuum. We used a red edge (i.e., the terminationwavelength of Band II) of 2.44 lm as the SpeXspectrograph loses sensitivity longward of 2.45 lm(Lindsay et al. 2016). Band centers and depths aremeasured three times using 3rd- through 5th-orderpolynomial fits to the lower one-half of Band I and tothe entirety of Band II. Final band center values aretaken as the average of the three polynomial fits. Errorsare calculated using a Monte-Carlo method, whichcreates 20,000 synthetic reflectance data points for eachpolynomial order, thereby generating 3 9 20,000measurements to obtain band center and depth errors.The final 1r center and depth errors are defined as theaverage of the three error measurements. For a detaileddescription of SARA, see Lindsay et al. (2015). Bandparameter results for untreated acapulcoite–lodranitemeteorites are presented in Table 7 and band results forEATG treated meteorites are listed in Table 8.

RESULTS

Petrography

Petrographic descriptions and mineral compositionsfor some (6 of 10) of the ALC samples have beenpublished by other workers. We report in thesupporting information brief petrographic descriptionsfor all 10 meteorites. The modal abundances of mineralphases for five samples (NWA 2871, LAR 06605, NWA5488, NWA 7312, and NWA 7674) were point countedat 1009 in polarized and reflected light using a stepinterval of 0.3 mm. We found average modalabundances (vol%) for acapulcoites (n = 3) of (1rstandard deviation in parentheses): olivine 27.8 (1.1),orthopyroxene 20.6 (13.1), clinopyroxene 8.1 (1.4),plagioclase 12.1 (1.7), FeNi metal 16.6 (11.9), troilite 5.9(4.0), chromite 0.6 (0.3), phosphates 0.6 (0.5), andlimonite (weathering products) 7.6 (11.4). For lodranites(n = 5), we found average modal abundances (vol%) of:olivine 36.4 (8.4), orthopyroxene 31.9 (16.2),clinopyroxene 5.4 (4.5), plagioclase 1.3 (1.8), FeNi metal13.5 (11.2), troilite 0.5 (0.8), chromite 0.3 (0.4),phosphates—trace, and limonite (weathering products)

Fig. 4. a) Cross-polarized photomicrograph mosaic of acapulcoite MET 01212. The fine-grained recrystallized, equigranulartexture, and abundant 120˚ triple junctions typical for acapulcoites are apparent in the thin section. Note the two different grainsizes within this rock. b) Cross-polarized photomicrograph mosaic of transitional acapulcoite–lodranite LEW 86220. Twolithologic textures are present within this specimen, a fine-grained region (~0.15 mm) with an acapulcoite-like texture andmineralogy (left side of image), and a coarser grained (0.5–2.0 mm) gabbroic-like region that is rich in plagioclase andclinopyroxene. The boundary between the two lithologies (red-dashed line) is not sharp and is difficult to demarcate.

Spectra and mineralogy of acapulcoites and lodranites 165

10.6 (10.3). Modal abundances for all 10 ALCmeteorites can be found in Table S1 of the supportinginformation. We have analyzed all samples for majorelement mineral chemistry (see the MineralCompositions section) and for spectral band parameters(see the Acapulcoite–Lodranite Spectral Results section)to enhance the interpretive power of these acapulcoite–lodranite meteorite analyses for associations withpotential asteroid spectral analogs. Of the 10 meteoritesfrom our sample suite, we only found a VIS+NIRreflectance spectra for one acapulcoite–lodranite (MAC88177) in the literature (Hiroi et al. 1993). We reportthe results of our VIS+NIR spectral analysis in theAcapulcoite–Lodranite Spectral Results section.

Mineral Compositions

OlivineOlivine core compositions are listed in Table 3.

Olivine fayalite composition (Fa = molar Fe/[molar

Mg+Fe] 9 100) ranges from Fa3.4 (NWA 7312) toFa14.0 (MAC 88177). Olivines within the lodranitegroup are generally more iron-rich than theacapulcoites. Olivine grains from several lodranitesexhibit reverse zoning (FeO decreases from core to rim).No zoning of olivine is observed in acapulcoites.

PyroxenesLow-Ca pyroxene core compositions are given in

Table 4, and high-Ca pyroxene core compositions arelisted in Table 5. Low-Ca pyroxene ferrosilite com-positions (Fs = molar Fe/[molar Mg+Fe+Ca] 9 100)show a broader range of Fs (Fs3.3 to Fs12.8), than forhigh-Ca pyroxene (Fs3.4 to Fs6.5). Similar to the olivinecompositions, pyroxenes in lodranites are more iron-richthan those in acapulcoites. Orthopyroxenes in thelodranites have higher average CaO and Al2O3 contentsthan those in the acapulcoites. The lodranites exhibitreverse CaO zoning (CaO decreases from core to rim) inorthopyroxene. Exsolution lamellae of orthopyroxene in

Table 3. Average major element compositions (wt% oxides above; atoms per formula unit below) for representativeolivine grains (basis of four oxygens).

Samplename

MET01212

Dhofar125

NWA725

NWA2871

LEW86220

LAR06605

MAC88177

NWA5488

NWA7312

NWA7674

# Grains n = 10 n = 10 n = 10 n = 10 n = 1 [3]c n = 12 n = 11 n = 10 [3]c n = 11 n = 11

SiO2 40.8 (2)a 40.4 (2) 40.7 (3) 40.4 (3) 40.9 (2) 40.2 (2) 39.6 (3) 39.9 (3) 41.7 (1) 40.5 (2)TiO2 n.d.b n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.Al2O3 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.04 (1) n.d.Cr2O3 0.03 (1) 0.08 (3) 0.04 (6) n.d. n.d. n.d. n.d. 0.06 (1) 0.25 (1) n.d.

FeO 8.63 (8) 8.50 (17) 6.23 (17) 8.57 (8) 7.42 (4) 11.6 (3) 13.5 (2) 10.4 (1) 3.36 (23) 10.3 (2)MnO 0.47 (2) 0.49 (2) 0.48 (1) 0.50 (2) 0.48 (3) 0.49 (1) 0.48 (1) 0.44 (1) 0.22 (1) 0.52 (2)MgO 50.3 (2) 50.4 (3) 51.9 (4) 50.4 (2) 51.2 (1) 47.8 (4) 46.3 (4) 48.4 (3) 54.0 (2) 49.1 (3)

CaO n.d. 0.06 (1) n.d. n.d. n.d. n.d. n.d. 0.04 (1) 0.14 (1) n.d.P2O5 n.d. n.d. n.d. n.d. 0.05 (1) n.d. n.d. n.d. n.d. n.d.∑ 100.3 (2) 99.9 (5) 99.4 (4) 100.0 (4) 100.2 (2) 100.1 (4) 99.9 (5) 99.3 (6) 99.8 (3) 100.5 (3)

Molar Fe/Mn; Fo = 100 9 Mg/(Mg+Fe); Fa = 100 9 Fe/(Mg+Fe)Fe/Mn 18.3 17.4 12.9 17.0 15.0 23.8 27.9 23.8 15.7 19.4Fo 91.2 91.3 93.7 91.3 92.5 88.0 86.0 89.2 96.6 89.5

Fa 8.8 8.7 6.3 8.7 7.5 12.0 14.0 10.8 3.4 10.5Atoms per formula unitSi 0.993 0.988 0.990 0.987 0.992 0.993 0.990 0.990 0.996 0.992Ti n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

Al n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.001 n.d.Cr 0.001 0.002 0.001 n.d. n.d. n.d. n.d. 0.001 0.005 n.d.Fe 0.176 0.174 0.127 0.175 0.150 0.240 0.282 0.217 0.067 0.211

Mn 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.009 0.004 0.011Mg 1.825 1.836 1.882 1.838 1.851 1.761 1.725 1.791 1.924 1.791Ca n.d. 0.002 n.d. n.d. n.d. n.d. n.d. 0.001 0.004 n.d.

P n.d. n.d. n.d. n.d. 0.001 n.d. n.d. n.d. n.d. n.d.∑ 3.005 3.012 3.010 3.010 3.004 3.004 3.007 3.009 3.001 3.005aUnits in parentheses represent one standard deviation of replicate analyses in terms of least units cited.bn.d. = not detected; <0.03 wt% for SiO2, MgO, TiO2, Al2O3, CaO, and Na2O; <0.04 wt% for K2O and Cr2O3; <0.05 wt% for FeO, MnO, and

P2O5.cNumbers within brackets represent number of spots analyzed per grain.

166 M. P. Lucas et al.

Table

4.Averagemajorelem

entcompositions(w

t%oxides

above;

atoms/form

ula

unitbelow)forlow-C

apyroxenegrains(basisofsixoxygens).

Sample

MET

01212

Dhofar125

NWA

725

NWA

2871

LEW

86220c

LAR

06605

MAC

88177

NWA

5488

NWA

7312

NWA

7674

#Grains

n=10

n=9

n=10

n=11

n=10

n=10

n=11

n=10[3]d

n=8

n=10

SiO

257.4

(2)a

57.4

(2)

57.1

(6)

56.8

(2)

57.7

(2)

56.7

(2)

56.4

(2)

57.4

(4)

57.9

(1)

56.5

(2)

TiO

20.22(2)

0.21(1)

0.16(2)

0.19(1)

0.23(2)

0.19(2)

0.14(1)

0.07(3)

0.15(1)

0.13(1)

Al 2O

30.35(2)

0.36(2)

0.22(4)

0.45(1)

0.32(2)

0.44(3)

0.59(5)

0.41(15)

1.31(2)

0.67(2)

Cr 2O

30.34(3)

0.45(2)

0.26(4)

0.59(5)

0.26(3)

0.56(7)

0.53(4)

0.37(19)

0.38(1)

0.95(2)

FeO

5.97(7)

5.49(11)

5.52(33)

6.33(5)

6.03(15)

7.93(7)

8.61(19)

6.97(23)

2.27(6)

7.37(7)

MnO

0.54(1)

0.55(2)

0.55(1)

0.56(1)

0.50(1)

0.54(1)

0.51(1)

0.47(1)

0.22(2)

0.54(1)

MgO

34.4

(2)

34.8

(2)

35.0

(4)

33.5

(2)

35.0

(3)

32.2

(3)

31.7

(3)

34.0

(4)

36.6

(1)

32.1

(1)

CaO

1.07(13)

1.08(13)

0.82(8)

1.68(10)

0.79(12)

1.77(24)

1.73(24)

0.95(31)

0.98(2)

2.13(2)

Na2O

n.d.b

0.04(1)

n.d.

0.05(1)

n.d.

0.06(2)

0.05(1)

n.d.

n.d.

0.08(1)

K2O

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

∑100.3

(2)

100.3

(3)

99.6

(6)

100.1

(3)

100.9

(3)

100.3

(3)

100.2

(3)

100.6

(5)

99.8

(2)

100.4

(2)

MolarFe/Mn;pyroxenemineralendmem

bers

Fe/Mn

10.7

9.8

9.9

11.1

11.6

14.6

16.7

14.5

10.6

13.4

Wo

2.0

2.0

1.5

3.2

1.5

3.3

3.3

1.8

1.8

4.0

En

89.3

90.0

90.5

87.6

89.9

84.9

83.9

88.1

94.9

85.0

Fs

8.7

8.0

8.0

9.3

8.7

11.7

12.8

10.1

3.3

11.0

Atomsper

form

ula

unit

Si

1.979

1.974

1.977

1.970

1.976

1.976

1.972

1.980

1.968

1.967

Ti

0.006

0.005

0.004

0.005

0.006

0.005

0.004

0.002

0.004

0.004

Al

0.014

0.015

0.009

0.018

0.013

0.018

0.024

0.017

0.052

0.027

Cr

0.010

0.012

0.007

0.016

0.007

0.015

0.015

0.010

0.010

0.026

Fe

0.172

0.158

0.160

0.184

0.173

0.231

0.252

0.201

0.065

0.215

Mn

0.016

0.016

0.016

0.017

0.015

0.016

0.015

0.014

0.006

0.016

Mg

1.768

1.785

1.806

1.733

1.788

1.672

1.655

1.746

1.856

1.666

Ca

0.040

0.040

0.030

0.062

0.029

0.066

0.065

0.035

0.036

0.079

Na

n.d.

0.003

n.d.

0.003

n.d.

0.004

0.004

n.d.

n.d.

0.006

Kn.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

n.d.

∑4.005

4.008

4.009

4.008

4.007

4.003

4.006

4.005

3.997

4.006

aUnitsin

parentheses

representonestandard

deviationofreplicate

analysesin

term

sofleast

unitscited.

bn.d.=notdetected;<0.03wt%

forSiO

2,MgO,TiO

2,Al 2O

3,CaO,andNa2O;<0.04wt%

forK

2O

andCr 2O

3;<0.05wt%

forFeO

,MnO,andP2O

5.

cAcapulcoite-likelithologyofLEW

86220.

dNumberswithin

bracketsrepresentnumber

ofspots

analyzedper

grain.

Spectra and mineralogy of acapulcoites and lodranites 167

clinopyroxene and vice versa are observed in somelodranites, but were not observed in the acapulcoites.High-Ca pyroxene is more abundant volumetricallywithin acapulcoites, and is less abundant or not detected(NWA 7312) in the lodranites. We analyzed high-Capyroxenes in both the acapulcoite-like and gabbroiclithologies of transitional acapulcoite–lodranite LEW86220 and their chemistries are remarkably similar(Table 5). High-Ca pyroxenes in both meteorite groupscontain significant amounts of Cr2O3 (1.10–1.73 wt%)and therefore are considered to be chromium diopsides(McCoy et al. 1997a).

Figure 5 shows the Fs of orthopyroxenes versusthe Fa of olivines for ALC meteorites. Orthopyroxene(i.e., low-Ca pyroxene) Fs values are well correlatedwith Fa of coexisting olivine. The mafic silicates of

ALC meteorites span a wide range of oxidation statesand plot between those of highly reduced enstatite (E)chondrites and H chondrites. Lodranites are generallymore iron-rich than acapulcoites and the two groupscluster in separate Fs versus Fa fields, although thereis some overlap between the groups (Fig. 5). NWA7312 has a significantly lower iron content comparedto the rest of the clan and plots in the E chondritefield.

PlagioclaseWe analyzed plagioclase in all of our acapulcoites

and as a major phase in only two lodranites (LAR06605 and NWA 7674). Plagioclase compositions arelisted in Table 6. Plagioclase is oligoclase incomposition and has relatively homogeneous K2O

Table 5. Average major element compositions (wt% oxides above; atoms/formula unit below) for representativehigh-Ca pyroxene grains (basis of six oxygen).

SampleMET01212

Dhofar125

NWA725

NWA2871

LEW86220c

LEW86220d

LAR06605

MAC88177 NWA 5488

NWA7674

# Grains n = 10 n = 10 n = 10 n = 10 n = 9 n = 2 [3]e n = 1 n = 4 n = 10 [3]e n = 6

SiO2 54.1 (2)a 54.0 (2) 53.8 (2) 54.0 (2) 54.1 (3) 54.2 (2) 54.0 53.5 (6) 54.1 (4) 53.7 (1)TiO2 0.58 (4) 0.62 (5) 0.56 (3) 0.45 (2) 0.73 (20) 0.72 (6) 0.37 0.30 (2) 0.11 (3) 0.33 (2)Al2O3 0.87 (11) 0.89 (10) 0.67 (2) 1.00 (15) 0.76 (10) 0.68 (12) 1.12 1.33 (6) 0.79 (25) 1.34 (3)Cr2O3 1.44 (13) 1.51 (16) 1.24 (3) 1.56 (11) 1.28 (7) 1.27 (5) 1.73 1.40 (12) 1.10 (21) 1.51 (4)

FeO 2.33 (22) 2.24 (15) 2.13 (13) 2.67 (21) 2.46 (7) 2.49 (19) 3.11 3.99 (59) 2.67 (41) 3.56 (21)MnO 0.30 (3) 0.32 (2) 0.27 (1) 0.32 (2) 0.29 (2) 0.32 (4) 0.30 0.33 (1) 0.26 (3) 0.35 (2)MgO 18.2 (5) 18.4 (2) 17.9 (1) 18.6 (5) 18.1 (3) 18.2 (4) 17.3 17.8 (7) 18.1 (4) 18.9 (4)

CaO 21.3 (7) 21.0 (6) 22.0 (3) 20.5 (9) 21.6 (2) 21.6 (5) 21.1 20.4 (9) 22.3 (5) 19.6 (5)Na2O 0.74 (4) 0.77 (5) 0.68 (2) 0.72 (3) 0.72 (3) 0.74 (4) 0.86 0.63 (4) 0.47 (3) 0.52 (4)K2O n.d.b n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

∑ 99.8 (4) 99.7 (2) 99.2 (2) 99.7 (2) 100.1 (4) 100.1 (2) 99.9 99.6 (5) 99.9 (5) 99.8 (3)Molar Fe/Mn; pyroxene mineral endmembersFe/Mn 7.6 6.8 7.7 8.1 8.4 7.8 10.6 11.9 10.2 10.0

Wo 44.0 43.4 45.4 42.3 44.4 44.2 44.3 42.3 45.0 40.2En 52.2 53.0 51.2 53.4 51.7 51.8 50.6 51.2 50.8 54.1Fs 3.8 3.6 3.4 4.3 3.9 4.0 5.1 6.5 4.2 5.7Atoms per formula unit

Si 1.966 1.963 1.967 1.962 1.964 1.965 1.968 1.956 1.970 1.952Ti 0.016 0.017 0.015 0.012 0.020 0.020 0.010 0.008 0.003 0.009Al 0.037 0.038 0.029 0.043 0.032 0.029 0.048 0.057 0.034 0.057

Cr 0.041 0.043 0.036 0.045 0.037 0.037 0.050 0.041 0.032 0.043Fe 0.071 0.068 0.065 0.081 0.075 0.076 0.095 0.122 0.081 0.108Mn 0.009 0.010 0.008 0.010 0.009 0.010 0.009 0.010 0.008 0.011

Mg 0.984 0.996 0.976 1.006 0.979 0.983 0.940 0.970 0.980 1.026Ca 0.829 0.816 0.864 0.797 0.841 0.838 0.824 0.801 0.869 0.763Na 0.052 0.054 0.049 0.051 0.051 0.052 0.060 0.045 0.033 0.037

K n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.∑ 4.005 4.005 4.009 4.007 4.008 4.010 4.004 4.010 4.010 4.006aUnits in parentheses represent one standard deviation of replicate analyses in terms of least units cited.bn.d. = not detected; <0.03 wt% for SiO2, MgO, TiO2, Al2O3, CaO, and Na2O; <0.04 wt% for K2O and Cr2O3; <0.05 wt% for FeO, MnO, and

P2O5.cAcapulcoite-like lithology of LEW 86220.dGabbroic-like lithology of LEW 86220.eNumbers within brackets represent number of spots analyzed per grain.

168 M. P. Lucas et al.

contents. Plagioclase chemistries range from An12.6 toAn20.0 (An = molar Ca/[molar Ca+Na+K] 9 100) forall meteorites except lodranite NWA 7674. The high-Cavalue (An36.8) for this lodranite was measured from asingle plagioclase grain enclosed within an olivine grain.McCoy et al. (1997a) found that lodranites with highFa content in olivine (~Fa7.5–13) tend to have very lowplagioclase modal abundances, and those lodraniteswith lower Fa content in olivine (~Fa<7.5) tend to havehigher plagioclase abundances (6.0–10.3 vol%). Alllodranites from this study have high iron contents inolivine (>Fa10.5). Plagioclase was not detected in onelodranite (NWA 7312) and a minor phase in two others(MAC 88177 and NWA 5488), so our results supportthis trend. Lodranite NWA 7312 does not follow thistrend, however, as plagioclase was not detected eventhough it has a very forsteritic olivine composition(Fa3.4). Plagioclase was not analyzed in MAC 88177,although we found a ~40 lm diameter SiO2-richfeldspathic glass melt inclusion within an olivine grain,

consistent with melt inclusions reported by Prinz et al.(1991) in the same meteorite.

Acapulcoite–Lodranite Spectral Results

The acapulcoite and lodranite spectra collectedduring this study reveal distinguishing absorptionfeatures controlled mainly by the Fe2+ content and themodal abundances of the mafic silicate minerals olivineand pyroxene. Olivine exhibits a broad absorption bandcentered at ~1 lm, and ortho- and clinopyroxenetypically display two strong absorptions centered at~0.9 lm and ~2.0 lm. Cloutis et al. (1986) showed thatolivine+orthopyroxene mixtures exhibit well-definedbehavior in band center versus band area ratio (BAR)parameter space. The analysis of our acapulcoite–lodranite spectra (Fig. 6) is therefore focused on olivineand orthopyroxene. These minerals are the dominantsilicate phases in these meteorites, and they are presentin quantities abundant enough to exhibit strong 1 and2 lm spectral absorptions. The results of our bandparameter analysis for untreated (i.e., before removal ofterrestrial alteration) sample spectra are given inTable 7, and the band parameter results of our samplespectra after removal of alteration via EATG treatmentare listed in Table 8. We also analyzed the spectralband parameters of the only recovered “falls” amongthe ALC, Acapulco, and Lodran, for spectralcomparison to our EATG-treated acapulcoite–lodranitemeteorite finds (Table 8). These falls show minimaleffects of terrestrial alteration and therefore providesuitable characteristic spectra of “fresh” acapulcoite–lodranite meteorites.

Two lodranites, NWA 5488 and NWA 7312,contain significant amounts of opx�cpx (see supportinginformation), but well-defined absorption bands areabsent in the 2 lm region (Fig. 6b). NWA 5488 isspectrally flat longward of a subtle ~1 lm absorptionfeature, and the spectrum of NWA 7312 displays agradual reddening in reflectance longward of the 1 lmregion. Spectra of materials containing pyroxene canexhibit a 1 lm band while also displaying little or noindication of a 2 lm absorption band (Cloutis et al.2010). This effect has been observed in shock-darkenedordinary chondrites (Britt and Pieters 1994) and hasbeen predicted from theoretical modeling (Moroz andArnold 1999). Shock darkening is caused by thedispersion of micron-scale FeNi metal and troilite (Brittand Pieters 1994) and may be a process affecting thesemeteorites. Without the presence of an ~2 lmabsorption, we cannot derive Band II centers and BARsfor these samples, so NWA 5488 and NWA 7312 arenot included in our spectral band parameter analysis.

Fig. 5. Iron contents of orthopyroxenes (Fs) and olivines (Fa)for 16 ALC meteorites (10 of our sample suite plus six fromthe literature; see below) in relation to those of enstatite (E)chondrites and ordinary chondrites (H, L, and LL). Increasingoxygen fugacity (fO2) trend shown with arrow. Lodranites(dashed box) are more iron-rich than acapulcoites (solid box)and the two groups plot in separate fields with some overlap.The E chondrite field is from Hiroi et al. (1993) and theordinary chondrite fields are from Dunn et al. (2010a). NWA7312 is highly reduced compared to the other ALC meteoritesand plots in the E chondrite field. Mineral compositions from:Acapulco (Zipel et al. 1995), Lodran (Bild and Wasson 1976),ALH 81261 and ALH 81187 (McCoy et al. 1996), EET 84302(Mittlefehldt et al. 1996), and GRA 95209 (McCoy et al.2006).

Spectra and mineralogy of acapulcoites and lodranites 169

Iron oxides/hydroxides (e.g., limonite) fromterrestrial weathering typically exhibit an absorptionedge near 0.55–0.6 lm, absorption bands centered near0.5 and 0.9 lm, and weaker absorptions associated withOH� may also be present in the 1.4 and 1.9 lm regions(Hiroi et al. 1993; Cloutis et al. 2010). The presence ofiron oxides/hydroxides can mask the characteristicspectral properties associated with the Fe2+ content andmodal abundances of olivine and pyroxene.

To compare the spectra of our weathered acapulcoite–lodranite powders with “cleaned” powder spectra, weplotted the Band I centers versus BAR for our ALCsamples before and after treatment with EATG (Fig. 7).From an inspection of Fig. 7 and a comparison of Band Iand BAR values for untreated samples (Table 7) andEATG-treated samples (Table 8), it is apparent thatterrestrial weathering has a minimal effect on band centerpositions. Measured Band I centers are identical or showminimal differences between the untreated meteoritespectra and the EATG-treated spectra. Band II centersalso show minimal differences, typically within 0.017 lmbetween untreated and treated samples. However, it isclear that BAR is significantly affected by the presencealteration products. The untreated, weathered samples

have lower measured BAR values than samples treatedwith EATG (Tables 7 and 8; Fig. 7). EATG-treatedsamples show deeper bands and therefore an increase inband area, with Band I exhibiting a smaller overallincrease in area due to the contribution of a ~0.9 lm bandfrom limonite in the untreated samples.

Although we demonstrate that BAR values increasewith removal of terrestrial alteration products, some ofour EATG-treated spectra still exhibit some noticeableeffects of weathering (Fig. 6). This is visible as a sharpabsorption edge near 0.6 lm for several meteorites (e.g.,LEW 86220 in Fig. 6c). An inflection of the steep red-slope near 0.5 lm for many of the spectra (e.g., Dhofar125 in Fig. 6a) is also most likely due to unremovedterrestrial alteration. Despite the presence of theseresidual spectral features due to weathering products,EATG-treated spectra are unmistakably shifted to higherBAR values, toward the “fresh” falls of the clan,Acapulco and Lodran.

As noted earlier, Gaffey et al. (1993) used Band Icenter versus BAR to characterize the olivine andpyroxene mineralogy and possible meteorite analogs ofstony asteroids of the S-complex, separating the S-complex into seven subtypes (S[I] through S[VII]). Later

Table 6. Average major element compositions (wt% oxides above; atoms/formula unit below) for representativeplagioclase grains (basis of eight oxygens).

Sample MET 01212 Dhofar 125 NWA 725 NWA 2871 LEW 86220c LEW 86220d LAR 06605 NWA 7674# Grains n = 10 n = 10 n = 2 [3]e n = 10 n = 10 n = 2 [3]e n = 2 n = 1

SiO2 63.8 (2)a 64.2 (4) 64.6 (6) 62.8 (6) 64.3 (4) 65.0 (11) 65.0 (5) 58.4

Al2O3 21.6 (2) 22.2 (2) 21.4 (3) 22.9 (5) 21.4 (3) 20.9 (9) 21.7 (9) 25.1FeO 0.14 (6) 0.41 (8) 0.62 (33) 0.22 (6) 0.26 (22) 0.16 (19) 0.38 (1) 0.39MgO n.d.b n.d. n.d. n.d. n.d. n.d. n.d. n.d.CaO 3.55 (11) 3.29 (3) 2.69 (4) 4.17 (46) 3.06 (20) 2.7 (9) 2.90 (11) 7.77

Na2O 9.18 (10) 9.47 (25) 9.37 (19) 8.92 (22) 9.52 (14) 9.7 (4) 8.5 (14) 7.21K2O 0.59 (4) 0.37 (5) 1.03 (19) 0.50 (7) 0.63 (8) 0.78 (26) 0.59 (4) 0.25∑ 98.9 (2) 100.0 (4) 99.7 (5) 99.5 (3) 99.2 (3) 99.2 (6) 99.1 (1) 99.2

Feldspar mineral endmembersOr 3.4 2.1 5.9 2.8 3.6 4.4 3.7 1.4Ab 79.6 82.1 81.2 77.2 81.9 83.0 81.0 61.8

An 17.0 15.8 12.9 20.0 14.5 12.6 15.3 36.8Atoms per formula unitSi 2.851 2.837 2.867 2.796 2.864 2.891 2.883 2.640

Al 1.138 1.159 1.121 1.202 1.124 1.095 1.134 1.337Fe 0.005 0.015 0.023 0.008 0.010 0.006 0.014 0.015Mg n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.Ca 0.170 0.156 0.128 0.199 0.146 0.128 0.138 0.376

Na 0.796 0.812 0.807 0.770 0.822 0.838 0.730 0.631K 0.034 0.021 0.059 0.028 0.036 0.044 0.034 0.014∑ 4.994 5.000 5.005 5.003 5.002 5.002 4.933 5.013aUnits in parentheses represent one standard deviation of replicate analyses in terms of least units cited.bn.d. = not detected; <0.03 wt% for SiO2, MgO, Al2O3, CaO, Na2O; <0.04 wt% for K2O3; <0.05 wt% for FeO.cAcapulcoite-like lithology of LEW 86220.dGabbroic-like lithology of LEW 86220.eNumbers within brackets represent number of spots analyzed per grain.

170 M. P. Lucas et al.

workers analyzed the spectral properties of meteoritesamples to delineate spectral zones in Band I versusBAR parameter space for the establishment of asteroid-meteorite connections with stony S-complex asteroids.Figure 8 shows Band I versus BAR of our EATG-treated acapulcoite–lodranite samples and the meteorite

falls Acapulco and Lodran (see Table 8) relative togeneralized spectral zones for five meteorite groups on amodified version (Burbine et al. 2001; Dunn et al. 2013)of the S-subtype plot.

The acapulcoite–lodranite meteorites we measuredhave Band I centers between 0.913 and 0.940 lm and

0.5 1.0 1.5 2.0 2.5

0.4

0.6

0.8

1.0

1.2

1.4

Nor

mal

ized

Ref

lect

ance

LEW 86220 (acapulcoite/lodranite)

(c) Transitional

0.5 1.0 1.5 2.0 2.5

1

2

3

4

Nor

mal

ized

Ref

lect

ance

Acapulco "fall"

Dhofar 125

NWA 725

MET 01212

NWA 2871

(a) Acapulcoites

0.5 1.0 1.5 2.0 2.5Wavelength (µm)

1

2

3

4

Nor

mal

ized

Ref

lect

ance

Lodran "fall"

LAR 06605

MAC 88177

NWA 5488

NWA 7312

NWA 7674

(b) Lodranites

Wavelength (µm)

Wavelength (µm)

Fig. 6. Reflectance spectra (0.3–2.55 lm) of ALC meteorites used in this study after removal of terrestrial alteration via EATGmethod. a) Acapulcoite spectra. b) Lodranite spectra. The ~2 lm feature due to pyroxene is absent in samples NWA 5488 andNWA 7312; therefore, these samples are not analyzed for spectral band parameters. c) Transitional meteorite LEW 86220spectrum. Spectra are normalized to a reflectance value of 1 at 0.55 lm and spectra in (a) and (b) are offset from each other by0.5 reflectance for clarity. (Color figure can be viewed at wileyonlinelibrary.com.)

Spectra and mineralogy of acapulcoites and lodranites 171

BARs between 0.46 and 0.99, and most of the samplescluster near the eponymous falls of the clan, Acapulcoand Lodran, in band parameter space. The samplesdisplay a restricted range of BAR values compared tothe generalized primitive achondrite spectral zone ofBurbine et al. (2001), and plot below the ordinarychondrite “boot” spectral region or overlap with thisregion. The lodranite samples (except for Lodran)display more overlap with the ordinary chondritespectral zone than do the acapulcoite samples (Fig. 8).High levels of shock can cause BAR to decrease, whileproducing no significant shift in Band I center (King1986; Cloutis et al. 2010). This is a plausibleexplanation for the low BAR value (0.529 � 0.002) forlodranite MAC 88177 relative to other samples (see“High shock” arrow in Fig. 8) as this meteorite isclassified as shock-stage S5 (Rubin 2007). AcapulociteNWA 2871 has the lowest BAR value (0.464 � 0.004)of all the samples and plots within the ureilite spectralzone (Cloutis et al. 2010). This low BAR value may bedue to incomplete weathering removal; this meteorite isone of the most weathered of our samples (weatheringgrade—W3). Excluding these two meteorites,acapulcoite–lodranite spectra share similar bandparameters, although there is some overlap with thebottom boundary of the ordinary chondrite spectralzone. The bottom portion of the ordinary chondriteboot is associated with the H chondrite group ofordinary chondrites (Dunn et al. 2010a). It maytherefore be challenging to discriminate betweenacapulcoite–lodranite and H chondrite meteoriteanalogs when comparing asteroid spectral bandparameters.

Olivine and Orthopyroxene Compositions/Spectral Band

Correlations

Examination of the FeO contents of theequilibrated olivine (Fa) and orthopyroxene (Fs) from

this study show that the data cluster into separate fieldsfor acapulcoites and lodranites (Fig. 5). Furthermore,the Fa and Fs for acapulcoites and lodranites are closeto their equilibrium relationships and appear to form alinear trend spanning the two groups. The relationshipbetween the measured Fa and Fs for 10 acapulcoite–lodranite meteorites for which we have spectral data(Table 8) is shown in Fig. 9 (axes inverted from Fig. 5).This relationship demonstrates that the Fa of olivineand the Fs of orthopyroxene in acapulcoite–lodraniteare well correlated and form a linear trend. We can usethis relationship to estimate the FeO content for one ofmafic silicates if the other is known. The Fa of olivinecan be calculated if the Fs of orthopyroxene is knownusing the linear equation:

Fa ¼ 1:100� Fs� 1:270: (1)

This linear relationship has an R2 of 0.86. Dunnet al. (2010a) showed that the Fa content of olivine andFs content of orthopyroxene in ordinary chondrites (H,L, and LL) correlate well with Band I center, withcorrelation coefficient (R2) values of 0.92 and 0.91,respectively. We found that the correlation betweenmafic silicate FeO contents and Band I center for ouracapulcoite–lodranite meteorites is poorly constrained,with R2 values of 0.17 for Fa, and a very lowcorrelation of 0.08 for Fs. The mafic silicate FeOcompositions of ordinary chondrites are mainlyinherited by the redox state of the solar nebula(Brearley and Jones 1998). The oxidation states of maficsilicates in acapulcoite–lodranite meteorites are lowerthan those of H chondrites, the most reduced group ofthe ordinary chondrites (Fig. 5). Partial melting ofchondritic material produces residues that becomeincreasingly magnesium-rich (FeO poor) with increasedpartial melting. Perhaps different processes controllingmafic silicate compositions and abundances as opposedto oxidation state (e.g., ordinary chondrites) is the

Table 7. Band parameter analysis for acapulcoite–lodranite meteorite samples (ground particulate <125 lm fraction;RELAB ID suffix “A” for untreated samples).

Meteorite Group RELAB ID BICBICerror

BI area(910�2)

BI areaerror(910�2) BIIC

BIICerror

BII area(910�2)

BII areaerror(910�2) BAR

BARerror

MET 01212 Acapulcoite MT-JPE-301-A 0.923 0.002 7.67 0.02 1.843 0.001 4.73 0.02 0.617 0.002Dhofar 125 Acapulcoite MT-JPE-304-A 0.920 0.001 3.88 0.02 1.840 0.002 2.14 0.02 0.552 0.005NWA 725 Acapulcoite MT-JPE-305-A 0.913 0.001 3.09 0.01 1.820 0.003 1.65 0.02 0.533 0.007NWA 2871 Acapulcoite MT-JPE-306-A 0.930 0.001 7.18 0.02 1.830 0.003 2.27 0.03 0.316 0.005

LAR 06605 Lodranite MT-JPE-300-A 0.923 0.001 12.37 0.02 1.853 0.001 7.70 0.04 0.623 0.004MAC 88177 Lodranite MT-JPE-302-A 0.940 0.001 13.04 0.02 1.863 0.002 5.64 0.02 0.432 0.002NWA 7674 Lodranite MT-JPE-310-A 0.930 0.001 6.83 0.02 1.853 0.001 3.88 0.02 0.569 0.004

LEW 86220 Acap/Lod MT-JPE-314-A 0.940 0.001 5.93 0.02 1.843 0.003 1.66 0.02 0.280 0.003

BIC = Band I center; BIIC = Band II center; BAR = band area ratio.

172 M. P. Lucas et al.

reason for the poor correlation of Fa and Fs with BandI center for ALC meteorites.

However, we found a correlation between the Fscontent of orthopyroxene within ALC meteorites andBand II center. The relationship between measured Fsand Band II center is shown in Fig. 10. Thisrelationship is described by a linear fit:

Fs ¼ 107:6� BIIC� 189:5; (2)

with Fs here being a spectrally derived value. Thislinear relationship has a R2 of 0.72. The correlationbetween Fs and Band II center allows for theestimation of the mol% Fs (spectrally derived) oforthopyroxene for S-complex asteroid spectra that areanalogous to acapulcoite–lodranite meteorites. Gaffeyet al. (2002) asserted that the presence of high-Capyroxene effects Band II position, which they suggestedis a linear function of the cpx/(opx+cpx) ratio in therange of values found in ordinary chondrites. Usingaccurate XRD-measured pyroxene abundances andBand II centers from a suite of ordinary chondrites,Dunn et al. (2010a) found no significant linearcorrelation (low R2 value of 0.15) between Band IIcenter and the relative pyroxene abundance. Regardless,perhaps two samples (acapulcoite Dhofar 125 andtransitional acapulcoite LEW 86220) that fall furthestfrom the opx Fs versus Band II center trend in Fig. 10can be explained by the higher abundance of cpx (seethe Petrography section) in acapulcoites (avg. 8.1 vol%)as compared to lodranites (avg. 5.4 vol%) and thepresence of significant amounts of cpx (roughly 20vol%) in the “gabbroic-like” lithology of LEW 86220(Fig. 4; see supporting information). Because thereappears to be a well-established correlation between Fa

and Fs in acapulcoite–lodranite meteorites (Fig. 9), theFa content of olivine from S-complex asteroid spectrathat are analogous to acapulcoite–lodranite meteoritescan be estimated from Equation 1. Figure 11a showsthat spectrally derived Fs values from Equation 2 arecorrelated with EPMA measured Fs. Interestingly, thetwo samples with the largest positive offsets from the1:1 derived to measured ratio line are the acapulcoiteDhofar 125 and the transitional acapulcoite LEW86220 discussed above. Higher cpx abundances inacapulcoites and in LEW 86220 as compared tolodranites may skew the Band II center with the resultthat these samples fall furthest from the trend.Figure 11b shows that derived Fa values usingEquation 1 are well correlated with EPMA measuredFa, with only slight offsets, except for a large positiveoffset for LEW 86220. Because the FeO contents of theacapulcoite group plot in a distinct field from that ofthe lodranite group (Fig. 5), the calculation of Fa andFs values from the band parameter analysis of S-complex asteroid spectra may allow the discriminationof ALC parent bodies.

DISCUSSION

The characteristics of acapulcoites and lodranitespresented here prompt interesting questions as to thenature of the ALPB, and the spectral and mineralogicproperties of the parent bodies of partially meltedasteroids. We discuss two considerations belowregarding these questions: the unexpectedly high FeOcontents of mafic silicates of the lodranite group relativeto the acapulcoite group (see the Puzzling Mafic SilicateFeO Contents of the Acapulcoite–Lodranite Clansection), and clues from the spectral properties of

Table 8. Band parameter analysis for acapulcoite–lodranite meteorite samples (ground particulate <125 lm fraction;RELAB ID suffix “T” for EATG treated samples; falls Acapulco and Lodran untreated). BAR0 is the band arearatio measured with a red edge of 2.44 lm, BARadj is the BAR adjusted to 2.50 lm.

Meteorite Group RELAB ID BIC

BIC

error

BI area

(910�2)

BI area

error

(910�2) BIIC

BIIC

error

BII

area

(910�2)

BII area

error

(910�2) BAR0

BAR0

error BARadj

Acapulco

(fall)

Acapulcoite TB-TJM-043 0.923 0.003 13.25 0.04 1.863 0.002 11.32 0.05 0.854 0.005 0.930

MET 01212 Acapulcoite MT-JPE-301-T 0.920 0.001 11.14 0.03 1.847 0.001 9.03 0.03 0.811 0.004 0.887

Dhofar 125 Acapulcoite MT-JPE-304-T 0.923 0.004 5.54 0.02 1.850 0.004 3.82 0.03 0.690 0.005 0.766

NWA 725 Acapulcoite MT-JPE-305-T 0.913 0.001 3.87 0.02 1.827 0.002 3.52 0.03 0.910 0.008 0.986

NWA 2871 Acapulcoite MT-JPE-306-T 0.930 0.001 8.01 0.03 1.840 0.003 3.72 0.03 0.464 0.004 0.540

Lodran (fall) Lodranite TB-TJM-041 0.920 0.002 13.95 0.05 1.880 0.001 12.10 0.03 0.868 0.004 0.944

LAR 06605 Lodranite MT-JPE-300-T 0.927 0.001 17.11 0.03 1.870 0.001 13.21 0.04 0.772 0.002 0.848

MAC 88177 Lodranite MT-JPE-302-T 0.940 0.001 18.13 0.03 1.873 0.001 9.59 0.03 0.529 0.002 0.605

NWA 7674 Lodranite MT-JPE-310-T 0.930 0.001 9.42 0.03 1.853 0.003 7.06 0.04 0.749 0.004 0.825

LEW 86220 Acap/Lod MT-JPE-314-T 0.937 0.003 6.77 0.03 1.860 0.004 4.81 0.04 0.710 0.007 0.786

BIC = Band I center; BIIC = Band II center; BAR = band area ratio.

Spectra and mineralogy of acapulcoites and lodranites 173

pyroxene-dominated mineralogies that help todistinguish between potential H chondrite andacapulcoite–lodranite parent bodies (see theDistinguishing H Chondrite and Acapulcoite–LodraniteClan Parent Asteroids section). Taken together, theseconsiderations provide evidence to decipher thepetrologic histories of some S-type asteroids, specificallywhether they are nebular (ordinary chondrites), or haveexperienced some degree of igneous processing(primitive achondrites).

0.0 0.5 1.0 1.5 2.0 2.5Band Area Ratio

0.90

0.91

0.92

0.93

0.94B

and

I Cen

ter

(µm

)Acapulcoite

Lodranite

Transitional (LEW 86220)

Solid symbols: untreatedOpen symbols: EATG treated

Tanzania Enstatite (Control)

Ol+En 50/50 wt% Mixture (Control)

Acapulco

Lodran

Fig. 7. Band I center versus BAR for ALC meteorites andnatural mineral control specimens. Acapulcoites (bluediamonds), lodranites (blue circles), transitional meteorite(black triangle), Tanzania enstatite (tan square), andolivine+enstatite 50/50 wt% mixture (green hexagon) bandparameters plotted for meteorite, and control spectra beforeweathering removal (solid symbols) and after weatheringremoval via EATG treatment (open symbols). Dashed tie linesconnect the same specimen. Band I centers are identical orminimally shifted for weathered versus EATG “cleaned”spectra. Weathered samples have reduced BAR valuescompared to EATG-treated samples. BARs shift towardhigher values with removal of terrestrial alteration. EATG-treated ALC meteorite samples shift toward the falls of theclan, Acapulco and Lodran (larger symbols). Errors are on theorder of the symbol size. (Color figure can be viewed at wileyonlinelibrary.com.)

Acapulco

Lodran

Primitive Achondrites

Increasing cpx

High shock

Fig. 8. Band I center versus BAR for EATG-treated ALCmeteorites shown relative to generalized spectral zones ofvarious stony meteorite groups. The olivine, ordinarychondrite, and basaltic achondrite zones are from Gaffey et al.(1993), the primitive achondrite zone is from Burbine et al.(2001), Dunn et al. (2013), and the ureilite zone is fromCloutis et al. (2010). The thick black curve indicates theolivine–orthopyroxene mixing line as in Fig. 1. Bandparameters of the only meteorite falls among the clan,Acapulco, and Lodran (larger symbols) are shown forcomparison. Changes in these spectral band parameters as afunction of increasing clinopyroxene content and shock effectsare indicated by arrows. The 1r errors for Band I center andBAR from Table 8 are plotted for each object (error istypically smaller than the symbol size). (Color figure can beviewed at wileyonlinelibrary.com.)

y = 1.100x - 1.270

R2 = 0.86

6 8 10 12 14Fs (mol% in orthopyroxene)

6

8

10

12

14

Fa

(mol

% in

oliv

ine)

LodraniteAcapulcoite

Transitional

Fig. 9. Iron contents of olivines (Fa) and orthopyroxenes (Fs)for 10 ALC meteorites for which we have spectral data. Therelationship between measured Fa and Fs forms a linear trendwith an R2 of 0.86.

1.75 1.80 1.85 1.90 1.956

8

10

12

14

16

18

1.75 1.80 1.85 1.90 1.95Band II center (µm)

6

8

10

12

14

16

18

Fs

(mol

% in

ort

hopy

roxe

ne)

LodraniteAcapulcoite

Transitional

Fs = 107.6x - 189.5R2 = 0.72

H chondrites

Fig. 10. Fs of orthopyroxene versus Band II center for the 10ALC meteorites from Fig. 9. This relationship demonstratesthat a correlation exists between the FeO content oforthopyroxene within these meteorites and Band II center.The linear fit has a R2 of 0.72. The gray box shows the fieldoccupied by H chondrites in this parameter space.

174 M. P. Lucas et al.

Puzzling Mafic Silicate FeO Contents of the

Acapulcoite–Lodranite Clan

Primitive achondrite meteorites have mineral andbulk compositions that are similar to chondrites in manyaspects, but have experienced varying degrees of partialmelting. The partial melting of chondritic material willproduce silicate partial melts that are enriched inplagioclase and pyroxene (basalts) that are distinctlymore ferroan than the solid residue. Therefore,progressive partial melting will result in residual maficsilicate minerals that become increasingly magnesian. TheALC is thought to originate from a single asteroidalparent body (ALPB) (Mittlefehldt et al. 1996; McCoyet al. 1997a). Acapulcoites have experienced very lowdegrees of partial melting and therefore have not lostsilicate melts, whereas many lodranites have experiencedhigher degrees of melting (up to ~20%) withtemperatures hot enough to have segregated basalticliquids. If both of these groups originated on a singleparent body from the same chondritic precursor material,then lodranites should have mafic silicates that are moremagnesian than acapulcoites. But in fact, most lodranitesfrom this study contain mafic silicates that are moreferroan than those within the acapulcoites (Tables 3 and4; Fig. 5), the converse of what is predicted from partialmelting. Our lodranite sample NWA 7312 is theexception, however, and contains highly magnesian(Fa3.4; Fs3.3) olivine and orthopyroxenes compared toour other lodranites and is similar in redox state to the Echondrite group (Fig. 5). The FeO contents of theresidual mafic silicates in NWA 7312 are lower than the

acapulcoite specimens in our sample suite and followthose that would be expected from progressive partialmelting. However, there is significant overlap of maficsilicate compositions in the range of Fa3–8 betweenacapulcoites–lodranites and other primitive achondritegroups, namely winonaites and IAB irons (Keil andMcCoy 2018). Unfortunately, more than 50% of ALCmeteorites currently have no supporting oxygen isotopedata (Keil and McCoy 2018), including NWA 7312.Within the range of mafic compositions for primitiveachondrites, supporting oxygen isotope data may be themost reliable method to differentiate between thesegroups. Additional definitive classifications for ALCmeteorites and for winonaites/IAB irons would help toconstrain mafic silicate compositional trends for thesegroups.

The reversed FeO content of mafic silicates for ALCmeteorites leads to the conclusion that the same chondriticprecursor material could not have produced both theacapulcoite and lodranite meteorite groups. If ALCmeteorites originated on the same parent body, theseprecursor materials must have been heterogeneous incomposition and were not homogenized by metamorphicor igneous processes (Mittlefehldt et al. 1996).

Distinguishing H Chondrite and Acapulcoite–LodraniteClan Parent Asteroids

Our equations to determine the mafic silicatecompositions of S-complex asteroid spectra may enablethe interpretation of the petrologic histories of someS-complex asteroids, specifically whether they are related

6 8 10 12 14Fs (measured)

6

8

10

12

14

Fs

(der

ived

)

LodraniteAcapulcoite

Transitional

(a)

6 8 10 12 14Fa (measured)

6

8

10

12

14

Fa

(der

ived

)

LodraniteAcapulcoite

Transitional

(b)

Fig. 11. Derived FeO contents versus measured FeO of mafic silicates for 10 ALC meteorites from Figs. 9 and 10. The soliddiagonal lines in each figure represent a 1:1 derived to measured ratio. a) Spectrally derived Fs values using Band II centercorrelate well with EPMA measured Fs for orthopyroxene with moderate positive offsets for one acapulcoite and the transitionalacapulcoite–lodranite LEW 86220, and slight negative offsets for several acapulcoites and lodranites. b) Derived Fa values usingEquation 1 versus EPMA measured Fa in olivine. Derived values correlate well with EPMA measured Fa, with only slightoffsets, except for a large positive offset for LEW 86220.

Spectra and mineralogy of acapulcoites and lodranites 175

to nebular, ordinary chondrite material, or if they haveexperienced partial melting, and are related to the ALC.However, one difficulty in discriminating between thesetwo distinct petrologic histories is that the spectralparameters in Band I versus BAR parameter space forour 10 acapulcoite–lodranite meteorites overlap withthose of the H chondrites. Dunn et al. (2010a) showedthat the different groups (H, L, and LL) of the ordinarychondrite class can be separated in Band I center versusBAR parameter space, with H chondrites occupying thelower portion of the OC “boot” meteorite spectral zone.Consequently, the overlap of these spectral parameterswould lead to uncertainty for the interpretation ofdetailed mineralogies for some S-complex asteroidspectra. Specifically, there would be ambiguity whether toestimate the FeO composition of mafic silicates with ourEquations 1 and 2, which are valid for acapulcoite–lodranite-like parent asteroids, or to derive mineralcompositions with the equations of Dunn et al. (2010a)that are valid to estimate the mineralogy of ordinarychondrite-like parent asteroids. To investigate thisoverlap beyond that of simply plotting spectralparameters relative to generalized meteorite spectralzones, we plotted the Band I centers and BARs of ouracapulcoite–lodranite samples versus those of 18 Hchondrites analyzed in Dunn et al. (2010a) (Fig. 12). Fortheir mineralogic calibration study of 48 ordinarychondrites, Dunn et al. (2010a) used a red edge of 2.5 lmfor Band II to derive BAR values. The choice of the rededge for Band II in the analysis of NIR spectra can leadto differences in Band II area, which will affect calculatedBARs (Lindsay et al. 2016). In order to directly compareasteroid spectra acquired with the SpeX spectrograph toour acapulcoite–lodranite meteorite spectral parameters,we analyzed our acapulcoite–lodranite spectra with a rededge of 2.44 lm. Therefore, to directly compare theBARs of our acapulcoite–lodranites to the H chondritesanalyzed in Dunn et al. (2010a), we adjusted our BARvalues to a red edge of 2.5 lm using the equation ofLindsay et al. (2016) with a slope coefficient for ordinarychondrites:

BARadj ¼ BAR0 � 1:265ðk0 � kREÞ; (3)

where BARadj is the adjusted BAR value, BAR0 is theBAR measured at the red edge of k0 (in this case2.44 lm), and kRE is the red edge wavelength beingadjusted to (in this case 2.5 lm). BARadj values arelisted in Table 8. Figure 12 shows that ALC meteoritescluster near the olivine+orthopyroxene mixing line ofCloutis et al. (1986), while H chondrites plot above themixing line. Band I center values for ALC meteoritesare on average lower than for H chondrites (averageBand I centers of 0.926 and 0.939 lm, respectively).

However, there is some overlap of BAR values betweenthe groups. It is apparent from Fig. 12 that Band Icenter versus BAR is not a definitive discriminatorbetween the ALC and H chondrites.

Relationships between spectral absorption bandcenters and pyroxene compositions Fs and Wo (mol%wollastonite; CaSiO3) were first examined with naturalpyroxenes by Adams (1974) and revisited by Cloutisand Gaffey (1991). These relationships were then furtherstudied using synthetic pyroxene samples by Klimaet al. (2007, 2011). Equations derived from thesespectral relationships have been applied to estimate thepyroxene compositions of olivine+pyroxene assemblagesfrom S-complex asteroid spectra (Gaffey et al. 2002).However, these mineralogic equations have been shownto overestimate (by 8–20 mol%) measured Fscompositions when applied to the spectra of ordinarychondrite-like mineral assemblages (McCoy et al. 2007),limiting the usefulness of these equations. Morerecently, absorption band centers have been used toestimate the pyroxene compositions of HED meteorites(Burbine et al. 2007, 2009), and band centers haveshown to be successful at separating the howardite,eucrite, and diogenites groups in plots of Band I centerversus Band II center parameter space (Burbine et al.2009; Moskovitz et al. 2010; Beck et al. 2011; McSweenet al. 2013).

0.0 0.5 1.0 1.5Band Area Ratio

0.900.92

0.94

0.96

0.98

1.00

1.02

1.04

Ban

d I C

ente

r (µ

m)

AcapulcoiteLodraniteTransitionalH chondrites (Dunn et al., 2010)

Lodran

AcapulcoH chondrites

Acapulcoite-Lodranite

ol+opx mixing line

L chondrites

LL chondrites

Fig. 12. Band I center versus BAR for ALC meteorites and Hchondrite meteorites (data from Dunn et al. 2010a). Generalizedspectral zone delineated for ordinary chondrites (boundaries forH, L, and LL ordinary chondrites indicated). We delineate a newgeneralized spectral zone for acapulcoite–lodranite clanmeteorites (blue rectangle). Larger symbols show eponymousmeteorite falls of the clan, Acapulco and Lodran. BAR values foracapulcoite–lodranite meteorites are adjusted for a red edge of2.50 lm for direct comparison to the H chondrite values (see theDistinguishing H Chondrite and Acapulcoite–Lodranite ClanParent Asteroids section). Acapulcoite–lodranite meteoritescluster near the olivine+orthopyroxene mixing line of Cloutiset al. (1986). H chondrites plot above the mixing line. Errors areon the order of the symbol size. (Color figure can be viewed atwileyonlinelibrary.com.)

176 M. P. Lucas et al.

To investigate if absorption band centers areeffective in discriminating between acapulcoite–lodraniteand H chondrites meteorites, we plotted Band I centerversus Band II center for our acapulcoite–lodranitemeteorites against the band centers of the same Hchondrites (Dunn et al. 2010a) shown in Fig. 12. Cloutiset al. (1986) showed that Band I and Band II centerpositions can be used as a gauge of FeO content inolivine and orthopyroxene, and that BAR is a usefulindicator of the relative modal abundance of theseminerals. The differences in FeO content of olivine (Fa)and orthopyroxene (Fs) in ALC meteorites (ranges ofFa3.5–15, Fs6.5–14.5) are lower than those of the Hchondrites (ranges of Fa16–20, Fs14.5–18) and thesedifferences in mafic silicate mineral chemistry (seeFig. 5) separate ALC meteorites from H chondrites inBand I versus Band II parameter space (Fig. 13). TheALC and the H chondrite group plot in distinct fields inthis parameter space, which enable the discrimination ofpotential asteroidal parent bodies between these groupsfrom band parameter analysis of S-complex asteroidspectra. Comparison of S-complex asteroid spectralparameters to these fields allows for the appropriatechoice of equations for the estimation of olivine andorthopyroxene FeO content. For example, if themeteorite analog spectral zone of an S-complex asteroidspectrum is ambiguous between the ALC and the Hchondrites in Band I center versus BAR parameter space(Fig. 12), an examination of Band I versus Band IIcenter parameter space should reveal the distinction. Ifband centers plot in the field defined by our acapulcoite–lodranite meteorites in Fig. 13, then Fs can be calculatedfrom Band II center using Equation 2, and then therelationship between Fa and Fs defined in Equation 1can be used to estimate Fa. If band centers plot in theregion defined by H chondrite meteorites in Fig. 13, thenFa and Fs can be calculated from the Band I centerusing equations valid for ordinary chondrites describedin Dunn et al. (2010a). The mineralogical distinctionbetween ALC meteorites and H chondrites can berevealed through spectral parameters, which enable thecharacterization of partially melted parent bodies fromnebular ordinary chondrite parent bodies.

SUMMARY

Spectral and mineralogical studies of meteoritesafford the best possibility of linking meteorite groupswith their parent asteroid(s) using solely ground-basedasteroid spectral observations. Using spectral properties,relationships between the mineralogy of stony meteoritesand S-complex asteroid surface mineralogy have beenwell established for nebular ordinary chondrites and forigneous basaltic achondrites. Here, we have analyzed the

spectral parameters and mafic silicate compositions of asuite of acapulcoite–lodranite clan meteorites toestablish asteroid–meteorite connections between thesepartial melt residues and their S-complex parentasteroids. Previously, the identification of potential S-complex parent asteroids for the acapulcoite–lodraniteclan was difficult in part due to the lack of high-qualityreflectance spectra of these relatively rare and usuallyweathered meteorites. To address this need, we applied amethod for the neutralization and removal of terrestrialweathering products. We show that the presence ofterrestrial weathering products in meteorite powdersmainly reduces measured BAR values and that the bandcenter positions of olivine and orthopyroxene are mostlyunaffected by the application of the EATG weatheringremoval method.

FeO mineral compositions of orthopyroxene (Fs)and coexisting olivine (Fa) are well correlated inacapulcoite–lodranites and form a linear trend in Faversus Fs parameter space that spans the two meteoritegroups of the clan. Lodranites are generally more iron-rich than acapulcoites (the opposite of the expectedpartial melting trend) and the two groups cluster inseparate Fs versus Fa fields, requiring that thechondritic source materials of the two groups weredifferent if they come from the same parent asteroid.Correlations between mineral chemistry and spectralparameters in our suite of 10 acapulcoite–lodranite clanmeteorites indicate that a well-defined relationship existsbetween Band II center and Fs content oforthopyroxene, thus the derived Fs content can be usedto estimate the Fa content of the coexisting olivine. Thesecorrelations can be used to calculate mineral compositionsfrom the spectra of S-complex parent bodies that arerelated to acapulcoite–lodranite meteorites. The spectralproperties of acapulcoite–lodranite clan meteorites cluster

1.75 1.80 1.85 1.90 1.95 2.00Band II center (µm)

0.85

0.90

0.95

1.00

1.05

1.10

Ban

d I c

ente

r (µ

m)

AcapulcoiteLodraniteTransitionalDunn et al. (2010) H chondrites

Acapulcoite-Lodranites

H Chondrites

Fig. 13. Band I center versus Band II center for ALC meteorites(this study) and H chondrite meteorites (data from Dunn et al.2010a). The ALC and the H chondrite group plot in distinct fieldsin this parameter space allowing for the discrimination ofpotential asteroidal parent bodies from the analysis of asteroidspectra. Errors are on the order of the symbol size. (Color figurecan be viewed at wileyonlinelibrary.com.)

Spectra and mineralogy of acapulcoites and lodranites 177

in Band I center versus BAR parameter space near thepreviously defined primitive achondrite spectral zone.However, acapulcoite–lodranite spectral parametersoverlap with those of H chondrites in this parameter space,blurring the ability to distinguish between nebularordinary chondrite parent bodies and parent bodies thatmay have experienced partial melting. Nevertheless, wefind that a plot of Band I versus Band II center parameterspace clearly distinguishes the acapulcoite–lodranite clanand the H chondrites. These data will enable more robustasteroid–meteorite connections for the identification of S-complex asteroids that may be parent bodies for primitiveachondrites.

Acknowledgments—RELAB is a multiuser facilitysupported by NASA programs such as SSERVI. We aregrateful to the MWG for supplying five Antarcticacapulcoite–lodranite meteorite samples. The authorsthank Christine Floss and Ed Cloutis for constructivereviews that significantly improved the manuscript. Weappreciate Colin Sumrall, Micah Jessup, Anthony Faiia,and Larry Taylor for providing laboratory space. We thankTim Diedesch for assistance with photomicrography. Weare obliged to Sean Lindsay for the development of theIDL-based SARA algorithm, and to Allan Patchen forassistance with EPMA. Support for this work was providedby NASA Earth and Space Sciences Fellowship grantNNX13AO69H to J.P.E. and M.P.L. M.P.L. is grateful tothe University of Tennessee Graduate School for an OscarR. Ashley Fellowship, and to the M.D. Anderson CancerCenter, Houston, TX.

Editorial Handling—Dr. Kevin Righter

REFERENCES

Adams J. B. 1974. Visible and near-infrared diffuse reflectancespectra of pyroxenes as applied to remote sensing of solidobjects in the solar system. Journal of GeophysicalResearch 79:4829–4836.

Beck P., Barrat J.-A., Grisolle F., Quirico E., Schmitt B.,Moynier F., Gillet P., and Beck C. 2011. NIR spectraltrends of HED meteorites: Can we discriminate betweenthe magmatic evolution, mechanical mixing andobservation geometry effects? Icarus 216:560–571.

Benedix G. K., McCoy T. J., Keil K., Bogard D. D., andGarrison D. H. 1998. A petrologic and isotopic study ofwinonaites: Evidence for early partial melting, brecciation,and metamorphism. Geochimica et Cosmochimica Acta62:2535–2553.

Bild R. W. and Wasson J. T. 1976. The Lodran meteorite andits relationship to the ureilites. Mineralogical Magazine40:721–735.

Brearley A. J. and Jones R. 1998. Chondritic meteorites. InPlanetary materials, edited by Papike J. J. Reviews inMineralogy and Geochemistry, vol. 36. Washington, D.C.:Mineralogical Society of America. pp. 3-281–3-296.

Britt D. T. and Pieters C. M. 1994. Darkening in black andgas-rich ordinary chondrites: The spectral effects ofopaque morphology and distribution. Geochimica etCosmochimica Acta 58:3905–3919.

Burbine T. H., McCoy T. J., Nittler L. R., and Bell J. F. III.2001. Could 433 Eros have a primitive achondriticcomposition (abstract #1860)? 32nd Lunar and PlanetaryScience Conference. CD-ROM.

Burbine T. H., McCoy T. J., Jarosewich E., and Sunshine J.M. 2003. Deriving asteroid mineralogies from reflectancespectra: Implications for the MUSES-C target asteroid.Antarctic Meteorite Research 16:185–195.

Burbine T. H., Buchanan P. C., and Binzel R. P. 2007.Deriving formulas from HED spectra for determining thepyroxene mineralogy of Vesta and vestoids (abstract#2117). 38th Lunar and Planetary Science Conference.CD-ROM.

Burbine T. H., Buchanan P. C., Dolkar T., and Binzel R. P.2009. Pyroxene mineralogies of near-Earth vestoids.Meteoritics & Planetary Science 44:1331–1341.

Burns R. G. 1993.Mineralogical applications of crystal field theory,2nd ed. New York: Cambridge University Press. 523 p.

Cheng A. F. 2002. Near Earth asteroid rendezvous: Missionsummary. In Asteroids III, edited by Bottke W. F. Jr.,Cellino A., Paolicchi P., and Binzel R. P. Tucson,Arizona: The University of Arizona Press. pp. 351–366.

Clayton R. N., Mayeda T. K., and Nagahara H. 1992.Oxygen isotope relationships among primitive achondrites(abstract). 23rd Lunar and Planetary Science Conference.p. 231.

Cloutis E. A. and Gaffey M. J. 1991. Pyroxene spectroscopyrevisited: Spectral-compositional correlations andrelationship to geothermometry. Journal of GeophysicalResearch 96:22,809–22,826.

Cloutis E. A., Gaffey M. J., Jackowski T. L., and Reed K. L.1986. Calibrations of phase abundance, composition, andparticle size distribution for olivine-orthopyroxenemixtures from reflectance spectra. Journal of GeophysicalResearch 91:11,641–11,653.

Cloutis E. A., Hudon P., Romanek C. S., Bishop J. L., ReddyV., Gaffey M. J., and Hardersen P. S. 2010. Spectralreflectance properties of ureilites. Meteoritics & PlanetaryScience 45:1668–1694.

Cloutis E. A., Sanchez J. A., Reddy V., Gaffey M. J., BinzelR. P., Burbine T. H., Hardersen P. S., Hiroi T., Lucey P.G., Sunshine J. M., and Tait K. T. 2015. Olivine-metalmixtures: Spectral reflectance properties and application toasteroid reflectance spectra. Icarus 252:39–82.

Cornish L. and Doyle A. 1984. Use of ethanolaminethioglycolate in the conservation of pyritized fossils.Palaenotology 27:421–424.

Dunn T. L., McCoy T. J., Sunshine J. M., and McSween H.Y. 2010a. A coordinated spectral, mineralogical andcompositional study of ordinary chondrites. Icarus208:789–797.

Dunn T. L., Cressey G., McSween H. Y., McCoy T. J., andBland P. A. 2010b. Analysis of ordinary chondrites usingpowder X-ray diffraction: 1. Modal mineral abundances.Meteoritics & Planetary Science 45:123–134.

Dunn T. L., Burbine T. H., Bottke W. F. Jr., and Clark J. P.2013. Mineralogies and source regions of near-Earthasteroids. Icarus 222:273–282.

Eugster O. and Lorenzetti S. 2005. Cosmic-ray exposure agesof four acapulcoites and two differentiated achondrites

178 M. P. Lucas et al.

and evidence for a two-layer structure of the acapulcoite/lodranite parent asteroid. Geochimica et CosmochimicaActa 69:2675–2685.

Eugster O., Herzog G. F., Marti K., and Caffee M. W. 2006.Irradiation records, cosmic ray exposure ages, and transfertimes of meteorites. In Meteorites and the early solar systemII, edited by Lauretta D. S. and McSween H. Y. Jr. Tucson,Arizona: The University of Arizona Press. pp. 829–851.

Floss C. 2000a. Complexities on the acapulcoite-lodranite parentbody: Evidence from trace element distributions in silicateminerals.Meteoritics & Planetary Science 35:1073–1085.

Floss C. 2000b. LEW 86220: Evidence for melt migration onthe acapulcoite-lodranite parent body (abstract #1277).31st Lunar and Planetary Science Conference. CD-ROM.

Gaffey M. J. 1997. Surface lithologic heterogeneity of asteroid4 Vesta. Icarus 127:130–157.

Gaffey M. J. 2010. Space weathering and the interpretation ofasteroid reflectance spectra. Icarus 209:564–574.

Gaffey M. J. and Gilbert S. L. 1998. Asteroid 6 Hebe: Theprobable parent body of the H-type ordinary chondritesand the IIE iron meteorites. Meteoritics & PlanetaryScience 33:1281–1295.

Gaffey M. J., Bell J. F., and Brown R. H. 1993. Mineralogicalvariations within the S-type asteroid class. Icarus 106:573–602.

Gaffey M. J., Cloutis E. A., Kelley M. S., and Reed K. L. 2002.Mineralogy of asteroids. In Asteroids III, edited by BottkeW. F. Jr., Cellino A., Paolicchi P., and Binzel R. P. Tucson,Arizona: The University of Arizona Press. pp. 183–204.

Greenwood R. C., Franchi I. A., Gibson J. M., and Benedix G.K. 2012. Oxygen isotope variation in primitive achondrites:The influence of primordial, asteroidal, and terrestrialprocesses. Geochimica et Cosmochimica Acta 94:146–163.

Hiroi T., Bell J. F., Takeda H., and Pieters C. M. 1993.Modeling of S-type asteroid spectra using primitiveachondrites and iron meteorites. Icarus 102:107–116.

Jarosewich E. 1990. Chemical analyses of meteorites: Acompilation of stony and iron meteorite analyses.Meteoritics 25:323–337.

Keil K. and McCoy T. J. 2018. Acapulcoite-lodranitemeteorites: Ultramafic asteroidal partial melt residues.Chemie Erde—Geochemistry 78:153–203.

King T. V. 1986. Contributions toward a quantitativeunderstanding of reflectance spectroscopy: Phyllosilicates,olivine and shocked materials. Ph.D. thesis, University ofHawaii, Honolulu, Hawaii, USA.

King T. V. and Ridley W. I. 1987. Relation of thespectroscopic reflectance of olivine to mineral chemistryand some remote sensing implications. Journal ofGeophysical Research 92:11,457–11,469.

Klima R. L., Pieters C. M., and Dyar M. D. 2007. Spectroscopyof synthetic Mg-Fe pyroxenes I: Spin-allowed and Spin-forbidden crystal field bands in the visible and near-infrared.Meteoritics & Planetary Science 42:235–253.

Klima R. L., Dyar M. D., and Pieters C. M. 2011. Near-infrared spectra of clinopyroxenes: Effects of calciumcontent and crystal structure. Meteoritics & PlanetaryScience 46:379–395.

Krot A. N., Keil K., Scott E. R. D., Goodrich C. A., andWeisberg M. K. 2007. Classification of meteorites. InTreatise on Geochemistry, edited by Holland H. D. andTurekian K. K. Amsterdam: Elsevier/Pergamon. pp. 1–52.

Lindsay S. S., Marchis F., Emery J. P., Enriquez J. E., andAssafin M. 2015. Composition, mineralogy, and porosity

of multiple asteroid systems from visible and near-infraredspectral data. Icarus 247:53–70.

Lindsay S. S., Dunn T. L., Emery J. P., and Bowles N. E.2016. The red edge problem in asteroid band parameteranalysis. Meteoritics & Planetary Science 51:806–817.

Lucas M. P., Emery J. P., Hiroi T., and Milliken R. E. 2016.Spectral properties of primitive achondrite meteorites:Establishing S-type asteroid–meteorite connections (abstract#3003). 47th Lunar and Planetary Science Conference.CD-ROM.

Marti K. and Graf T. 1992. Cosmic-ray exposure history ofordinary chondrites. Annual Review of Earth and PlanetarySciences 20:221–243.

Martins Z., Hofmann B. A., Gnos E., Greenwood R. C.,Verchovsky A., Franchi I. A., Jull A. G. T., Botta O.,Glavin D. P., Dworkin J. P., and Ehrenfreund P. 2007.Amino acid composition, petrology, geochemistry, 14Cterrestrial age, and oxygen isotopes of the Shis�r 033 CRchondrite. Meteoritics & Planetary Science 42:1581–1595.

McCord T. B., Adams J. B., and Johnson T. V. 1970.Asteroid Vesta: Spectral reflectivity and compositionalimplications. Science 168:1445–1447.

McCoy T. J., Keil K., Clayton R. N., Mayeda T. K., BogardD. D., Garrison D. H., Huss G. R., Hutcheon I. D., andWieler R. 1996. A petrologic, chemical, and isotopic studyof Monument Draw and comparison with otheracapulcoites: Evidence for formation by incipient partialmelting. Geochimica et Cosmochimica Acta 60:2681–2708.

McCoy T. J., Keil K., Clayton R. N., Mayeda T. K., BogardD. D., Garrison D. H., and Wieler R. 1997a. A petrologicand isotopic study of lodranites: Evidence for earlyformation as partial melt residues from heterogeneousprecursors. Geochimica et Cosmochimica Acta 61:623–637.

McCoy T. J., Keil K., Muenow D. W., and Wilson L. 1997b.Partial melting and melt migration in the acapulcoite–lodranite parent body. Geochimica et Cosmochimica Acta61:639–650.

McCoy T. J., Nittler L. R., Burbine T. H., Trombka J. I.,Clark P. E., and Murphy M. E. 2000. Anatomy of apartially differentiated asteroid: A “NEAR”-sighted viewof acapulcoites and lodranites. Icarus 148:29–36.

McCoy T. J., Burbine T. H., McFadden L. A., Starr R. D.,Gaffey M. J., Nittler L. R., Evans L. G., Izenberg N.,Lucey P. G., Trombka J. I., Bell J. F. III, Clark B. E.,Clark P. E., Squyres S. W., Chapman C. R., Boynton W.V., and Veverka J. 2001. The composition of 433 Eros: Amineralogical-chemical synthesis. Meteoritics & PlanetaryScience 36:1661–1672.

McCoy T. J., Carlson W. D., Nittler L. R., Stroud R. M.,Bogard D. D., and Garrison D. H. 2006. Graves Nunataks95209: A snapshot of metal segregation and coreformation. Geochimica et Cosmochimica Acta 70:516–531.

McCoy T. J., Corrigan C. M., Sunshine J. M., Bus S. J., andGale A. 2007. Does asteroid spectroscopy provide evidencefor widespread partial melting of asteroids?: II. Pyroxenecompositions (abstract #1631). 38th Lunar and PlanetaryScience Conference. CD-ROM.

McSween H. Y., Binzel R. P., De Sanctis M. C., AmmannitoE., Prettyman T. H., Beck A. W., Reddy V., Le Corre L.,Gaffey M. J., McCord T. B., Raymond C. A., Russell C.T., and The Dawn Science Team. 2013. Dawn; the Vesta-HED connection; and the geologic context for eucrites,diogenites, and howardites. Meteoritics & PlanetaryScience 48:2090–2104.

Spectra and mineralogy of acapulcoites and lodranites 179

Mittlefehldt D. W., Lindstrom M. M., Bogard D. D., GarrisonD. H., and Field S. W. 1996. Acapulco- and Lodran-likeachondrites: Petrology, geochemistry, chronology, andorigin. Geochimica et Cosmochimica Acta 60:867–882.

Mittlefehldt D. W., Bogard D. D., Berkley J. L., and GarrisonD. H. 2003. Brachinites: Igneous rocks from a differentiatedasteroid. Meteoritics & Planetary Science 38:1601–1625.

Moroz L. V. and Arnold G. 1999. Influence of neutralcomponents on relative band contrasts in reflectancespectra of intimate mixtures: Implications for remotesensing 1. Nonlinear mixing modeling. Journal ofGeophysical Research 104:14,109–14,121.

Moskovitz N. A., Willman M., Burbine T. H., Binzel R. P.,and Bus S. J. 2010. A spectroscopic comparison of HEDmeteorites and V-type asteroids in the inner Main Belt.Icarus 208:773–788.

Nehru C. E., Prinz M., Weisberg M. K., Ebihara M. E.,Clayton R. N., and Mayeda T. K. 1996. A new brachiniteand petrogenesis of the group. Proceedings, 27th Lunarand Planetary Science Conference. p. 943.

Patzer A., Schultz L., and Franke L. 2003. New noble gasdata of primitive and differentiated achondrites includingNorthwest Africa 011 and Tafassasset. Meteoritics &Planetary Science 38:1485–1497.

Patzer A., Hill D. H., and Boynton W. V. 2004. Evolution andclassification of acapulcoites and lodranites from a chemicalpoint of view. Meteoritics & Planetary Science 39:61–85.

Pieters C. M. and Hiroi T. 2004. RELAB (ReflectanceExperiment Laboratory): A NASA multiuser spectroscopyfacility (abstract #1720). 35th Lunar and Planetary ScienceConference. CD-ROM.

Prinz M., Nehru C. E., Delaney J. S., and Weisberg M. 1983.Silicates in IAB and IIICD irons, winonaites, lodranites, andBrachina: A primitive and modified primitive group (abstract).14th Lunar and Planetary Science Conference. p. 616.

Prinz M., Chatterjee N., Weisberg M. K., Clayton R. N., andMayeda T. K. 1991. MAC 88177: A new type ofachondrite? (abstract). 22nd Lunar and Planetary ScienceConference. p. 1099.

Reddy V., Nathues A., Gaffey M. J., and Schaeff S. 2011.Mineralogical characterization of potential targets for theASTEX mission scenario. Planetary and Space Science59:772–778.

Rubin A. E. 2007. Petrogenesis of acapulcoites and lodranites:A shock melting model. Geochimica et Cosmochimica Acta71:2383–2401.

Russell C. T., McSween H. Y., Jaumann R., and Raymond C.A. 2015. The dawn mission to Vesta and Ceres. InAsteroids IV, edited by Michel P., DeMeo F. E., andBottke W. F. Tucson, Arizona: The University of ArizonaPress. pp. 419–432.

Sanchez J. A., Reddy V., Kelley M. S., Cloutis E. A., Bottke W.F., Nesvorn�y D., Lucas M. P., Hardersen P. S., Gaffey M.J., Abell P. A., and Le Corre L. 2014. Olivine-dominatedasteroids: Mineralogy and origin. Icarus 228:288–300.

Singletary S. and Grove T. L. 2006. Experimental constraintson ureilite petrogenesis. Geochimica et Cosmochimica Acta70:1291–1308.

Sunshine J. M., Bus S. J., Corrigan C. M., McCoy T. J., andBurbine T. H. 2007. Olivine-dominated asteroids andmeteorites: Distinguishing nebular and igneous histories.Meteoritics & Planetary Science 42:155–170.

Thomas C. A. and Binzel R. P. 2010. Identifying meteoritesource regions through near-Earth object spectroscopy.Icarus 205:419–429.

Warren P. H. and Kallemeyn G. W. 1992. Explosive volcanismand the graphite-oxygen fugacity buffer on the parentasteroid(s) of the ureilite meteorites. Icarus 100:110–126.

Zipel J., Palme H., Kennedy A. K., and Hutcheon I. D. 1995.Chemical composition and origin of the Acapulcometeorite. Geochimica et Cosmochimica Acta 59:3607–3627.

SUPPORTING INFORMATION

Additional supporting information may be found inthe online version of this article:

Fig. S1. Backscattered electron (BSE) images ofportions of acapulcoite polished sections: a) MET01212’s granular, metal-rich texture displays well-defined metal blebs, metal within the grain interstices,and metallic veins that crosscut the silicates. b) Dhofar125’s equigranular, recrystallized texture, and abundant120° triple junctions are readily visible in this image.Most grain interstices are filled with FeNi metal. c)NWA 725 is very fine-grained (~0.1 mm) for anacapulcoite. This rock has a granular texture withabundant 120˚ triple junctions and metal blebs. d) InNWA 2871, grain interstices are filled with metal andmetallic veins crosscut the silicates, and FeNi metalappears to have undergone melt migration in grainsinterconnected with vein channels.

Fig. S2. Backscattered electron (BSE) images ofportions of lodranite polished sections: a) MAC 88177

contains clinopyroxene as an interstitial phase thatdisplays exsolution lamellae of orthopyroxene andinclusions of metal and troilite. b) NWA 5488 has olivinegrains that contain numerous blebs of FeNi metal and/ortroilite and are typically crosscut by metallic veins.Clinopyroxene contains exsolution lamellae oforthopyroxene and discontinuous veinlets of metal andtroilite. c) NWA 7312 has a crystalline texture withabundant 120˚ triple junctions, coarse grain-size, olivineand orthopyroxene make up most of the mineralogy ofthis meteorite, plagioclase was not detected in our polishedsection, and some orthopyroxene grains contain bleb-likeinclusions of olivine. d) NWA 7674’s olivine grains containnumerous blebs of FeNi metal and/or troilite and aretypically crosscut by metallic veins, with metal ubiquitousin the grain interstices.

Table S1. Modal abundances of phases inacapulcoites and lodranites (vol%).

Data S1. Acapulcoite petrographic descriptions.Data S2. Lodranite petrographic descriptions.

180 M. P. Lucas et al.