Brimhall Et Al-2001

0361-0128/01/3140/401-20 $6.00 401

IntroductionWhere the telescope ends, the microscope begins.

Which of the two has the grander view?

Les Miserables by Victor Hugo, Book III, Chapter 3.

TYPICALLY, exploration geologists apply knowledge of rela-tively large scale crustal and district-scale genetic processes inthe search for ore deposits. However, mining of large-tonnage,low-grade ores has undergone a technological revolution,

which has required adding yet a new microscopic focus toeconomic geology that better bridges the scientific descrip-tion of orebodies with growing demands on mineralogical un-derstanding applied to metallurgy. Besides a traditional orereserve, todays data base requirements include the distribu-tion of mineral species in order to correlate the physical,chemical, and metallurgical behavior of the different types ofores and gangue with their response to treatment. This data-base is the primary input in development of performance andproduction cost models, which must be optimized as priceschange unpredictably and too often descend to remarkably

Atacamite Inclusions in Rock-Forming Feldspars and Copper-Bearing Smectites fromthe Radomiro Tomic Mine, Chile: Copper-Insoluble Mineral Occurrences

GEORGE H BRIMHALL,

Department of Earth and Planetary Science and the Earth Resource Center, University of California at Berkeley, Berkeley, California 94720-4767

BEATRIZ LEVI,Kallforsvgen 8, SE-12432 Bandhagen, Sweden

JAN OLOV NYSTRM,Swedish Museum of Natural History, Box 50007, SE-10405 Stockholm, Sweden

AND ENRIQUE TIDY F.Exploration Division, CODELCO, Santiago, Chile

AbstractRecovery of copper from the Radomiro Tomic deposit in Chile employs innovative large-scale hydrometal-

lurgical extraction methods, with sulfuric acid used for heap leaching. Here, we report on improvements in un-derstanding mineral components of copper oxide ores, which are insoluble in acid leach solutions, in order toadvance leaching technology. Using optical, electron microprobe, X-ray diffraction, energy dispersive spec-troscopy, and least squares inversion methods, two different occluded mineral occurrences were discovered inoxide ores, which present two distinct metallurgical challenges for heap leaching but offer important new op-portunities in process mineralogy. First, in a high chloride to sulfate ratio, feldspar-stable portion of the ore-body at Radomiro Tomic, disseminated atacamite (Cu4Cl2(OH)6) inclusions, 1 to 8 m in diameter, occur inmicrofractured feldspars and biotites. These account for the acid-insoluble fraction, often as high as 30 percentof the total contained copper. About 70 percent of the total copper occurs generally as atacamite in cracks,which upon crushing is exposed along surfaces of the rock fragments and is hence soluble. In contrast, in moreintensely hydrothermally altered, feldspar destructive, weak argillic alteration zones, insoluble copper existsmore probably within the crystal structure of well-crystallized saponitic smectite clays in nonexchangeable, oc-tahedral crystallographic sites. The two different forms of applied acid-insoluble components are products ofcontrasting microchemical environments, one in a reactive potassic alteration gangue and the second in a non-reactive gangue. In both cases, however, the nature of hydrothermal alteration, though pervasive, was relativelyweak and left intact rock mineral buffers capable of neutralizing acidic and reducing fluids. Hence, the insol-uble components with respect to the applied sulfuric acid reflect strong wall-rock mineral assemblage controlon the behavior of copper on all scales: the macrofield zonal scale, the microscale of individual crystals incracks, and the atomic scale of octahedral sites in the clay structure.

The metallurgical implications of these atacamite microinclusions in rock-forming minerals and structurallybound copper in smectites are different. In the former case, exposing ore inclusions on fragment surfaceswould require extremely costly grinding to a grain size well below 400 mesh. Secondly, chemical extraction ofmicroinclusion copper from feldspar host phases would entail high acid consumption, as hydrolysis of feldsparsby chemical reaction with applied acid would occur causing neutralization. In the case of Cu smectites, sincethe copper occurs in structurally bound, nonexchangeable sites, processes to remove the copper via ionexchange or acidic leach cannot work. Alternatives to ion exchange could involve destroying the smectite hostmineral, perhaps by its conversion to kaolinite that may only incorporate copper or other divalent cations innegligible amounts.

Economic GeologyVol. 96, 2001, pp. 401420

Corresponding author: e-mail, [email protected]

low levels. Translation of ore characterization into practical,cost-effective advancements in metallurgy is the basis ofprocess mineralogy, a relatively new discipline which is grow-ing rapidly in importance and becoming a vital new part ofeconomic geology. In the heap leaching of oxide copper ores,recovery of billion dollar capital costs of mine developmentand long-term profitability demand simultaneous maximiza-tion of copper recovery while minimizing acid consumptionand crushing costs. These competing demands require com-prehensive knowledge of ore mineral distribution of both theeasily recovered and the occluded ore mineral componentsthat presently are lost.

Porphyry copper deposits present a broad array of geomet-allurgical problems with respect to process mineralogy. In tra-ditional sulfide ores, extraction via froth flotation followingfine grinding to the feasible economic limit, ore mineralgrains remain interlocked with gangue to some extent andlower the concentrate grade by dilution while raising the oremetal grade of mill tailings. In comparison, acid heap leachextraction of copper from oxide ores presents its own uniquemineralogical barriers, which we illustrate here with the ex-amples of two relatively new ore types which can yield a lowcopper recovery: atacamite ore and copper clay ore. Ourstudy was carried out during the development and early pro-duction phases of the new Radomiro Tomic mine, which is lo-cated 5 km north of Chuquicamata in the Atacama Desert ofnorthern Chile (Cuadra and Rojas, 2001). It is largely basedon samples from the upper part of the oxidation zone.

The copper mineralization of Radomiro Tomic is hosted bya granodioritic to monzogranitic porphyry intrusion, theChuqui porphyry (Cuadra and Rojas, 2001), and extends to alesser extent upward into the overlying basal alluvial gravelsin certain mineralized paleochannels (Brimhall and Arcuri,1998). The 150- to 200-m-thick oxidation zone is complex anddifficult to subdivide into well-defined zones because the fourmain copper-bearing phases, atacamite, clay, and chrysocolla,plus Cu wad, occur in highly variable proportions at differentscales. X-ray diffractometry suggests that the polymorphparatacamite is widespread, associated with atacamite in theupper part of the oxidation zone, but since it is chemically in-distinguishible from the latter it is not treated as a separatemineral in this study. The extent of alteration of the host rockis also highly variable, from partially altered porphyry withcopper-bearing phases in fractures to an almost completelydegraded rock. The primary minerals are phenocrysts ofquartz, plagioclase, perthitic K feldspar, and biotite (bothphenocrysts and pseudomorphic replacement of hornblende)in a groundmass of quartz and saccharoidal K feldspar; thetwo latter also occur as veinlets.

Geologic SettingIn a schematic vertical cross section (Fig. 1), copper-bearing

clay minerals tend to characterize the upper, very heteroge-neous part of the oxidation zone, whereas the lower zone gen-erally is dominated by atacamite mineralization. The copper-bearing minerals form veinlets and coat fractures. Atacamite

402 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 402

FIG. 1. East-west vertical cross section along 10,600 N, looking north through the Radomiro Tomic copper deposit. Ore-type zones are shown. 200-m grid shows map scale.

EW SECTION 10,600N

Box 1 shows provenance of composite samples 1 through6 for the atacamite study and samples rt-2915 a, b, d and f for the Cu-bearing clays study

Gravels

Mixed ore

Chrysocolla-atacamitesmectite ore

Enriched ore

Chrysocolla-smectite ore(minor atacamite)

Primary ore

Atacamite-dominatedoxide ore

also occurs as dissemination and microfractures in feldspar,and locally, clays constitute an incipient to complete replace-ment of feldspar. Chrysocolla is preferentially associated withthe clays and gypsum. The clay minerals, which occur in dif-ferent shades and hues of yellowish-green, are clearly super-gene. They overprint bluish-green atacamite mineralizationin the upper zone and penetrate deeply into the lower zonealong faults and fractures. On the other hand, there are largeblocks rich in atacamite and poor in clays in the upper zone,especially in sectors relatively unaffected by fracturing orfaulting. Dissolution and reprecipitation of atacamite in clay-dominated domains further complicates the picture.

Analytical StrategyThe main purpose of this study was to identify and charac-

terize the mineralogy of ores, which, when leach solutions areapplied, are insoluble in the sulfuric acid solutions. A practi-cal visual ore classification scheme used at Radomiro Tomicidentifies two ore types that were investigated here: atacamiteore and copper clay ore. Toward this end, an analyticalmethod that was both efficient and definitive had to be de-vised in order to be cost effective and to provide statisticallyvalid results.

Since the nature of the occluded copper fraction in theatacamite zones was not at all obvious, we developed an iter-ative strategy that began with automated gridded point sam-pling for copper, using the electron microprobe on well-polished samples. Then, using the results as guides as to whichhost minerals to study in more detail, we were able to confirmthe conclusions using high-magnification optical microscopy.

Another strategy using XRD was implemented for the clay min-erals because they did not take a good polish, thus prohibitingboth electron mircoprobe or polished section petrographic ap-proaches. It was known from chemical analysis of clay fractionsthat the clays could contain considerable amounts of copper.Thus, other objectives of this study were to identify the clayminerals, to determine how much copper they contain, and toestablish if the copper in the clay is structurally bound or is dueto the presence of submicroscopic grains of copper minerals.

Samples

Six representative samples of atacamite ore were preparedfrom blast hole samples from the 2,915-m bench (Fig. 1, box1). The numbers of the samples are 1-2915-3, 2-2915-4/5, 3-2915-3, 4-2915-6/7, 5-2915-6, and 6-2915-7/6 and are re-ferred to here, respectively, as samples 1 through 6. Chemicalanalysis and assaying was done for total copper (Cutotal; range= 0.290.92 wt %), sulfuric acid soluble copper (CuS; 0.190.82 wt %), chloride (Cl; 0.140.29 wt %), iron (Fe; 0.551.07wt %), and carbonate (CO3; 0.120.47 wt %).

The clay minerals were studied in seven samples; three ofthem were collected from the 2,915-m bench near the blasthole sampling locations in Figure 1. The sample numbers areRT-2915-A, RT-2915-B, and RT-2915-D; they are clay domi-nated in spite of the vicinity to the blast hole samples (nos. 1,2, 4, and 6, respectively), which emphasizes the spatial vari-ability of the mineralization. Additional samples rich in clayminerals came from two exploration tunnels shown in Figure2 (samples RT-13007, RT-13014, and RT-13923) and a drillcore (DDH-3758; 151.9-m level).

RADOMIRO TOMIC MINE, CHILE: Cu-INSOLUBLE MINERAL OCCURRENCES 403

0361-0128/98/000/000-00 $6.00 403

FIG. 2. East-west vertical cross section along 10,700 N, looking north through the Radomiro Tomic copper deposit. 200-m grid shows map scale.

EW SECTION 10,700N

Box 1 shows location of sample DDH-3758 151.9m (smectite)Box 2 shows location of sample RT-13007 (smectite)Box 3 shows location of samples RT-13014 and RT-13923 (smectite)

Gravels

Mixed ore

Chrysocolla-atacamitesmectite ore

Enriched ore

Chrysocolla-smectite ore(minor atacamite)

Primary ore

Atacamite-dominatedoxide ore

Atacamite Ore

Sample preparation

The blast hole samples consist of granular material createdfrom intact rocks in the drilling process. In order to be able toprepare mounts for examination by microscopic and micro-chemical methods that require polishing, we screened eachsample into four different size ranges, using 20-, 100-, and140-mesh sieves. In this fashion, treatment of grains of aboutthe same diameter allowed us to make high-quality polishedepoxy grain mounts for use on the electron microprobe.Without the screen classification step, both large and smallgrains would occur together, which would result in overesti-mation or underestimation of grains of different size. Fourseparate epoxy plugs of a 2.54-cm diameter were cast, eachcontaining the same size fraction of all samples.

Polished sections were optically studied for each of the sizefractions. Very few sulfide minerals were identified evenunder 400 or 1,000 magnification, disproving the previous as-sertion that insoluble copper was in the form of sulfides. Weconclude instead that copper sulfide minerals are an insignif-icant contribution to the acid-insoluble fraction of all of thesesamples.

Analytical methods

A Cameca SX-50 electron microprobe at the University ofCalifornia, Berkeley, was used to study the samples, numbers1 to 6, using a beam diameter of about 5 m and an acceler-ating voltage of 15 kV with a sample current of 20 nA. The fol-lowing elements were analyzed: copper (Cu), iron (Fe), man-ganese (Mn), oxygen (O), chlorine (Cl), sodium (Na),potassium (K), aluminum (Al), silicon (Si), magnesium (Mg),calcium (Ca), and cobalt (Co). Counting times varied with 2,5, or 10 s on peak with an equal time-off peak dependingupon the element. For copper, the analytical detection levelis about 0.064 wt percent Cu for a 95 percent confidencelevel, 0.049 wt percent Cu for 90 percent, 0.035 wt percentCu for 80 percent, and 0.022 wt percent for a 60 percent con-fidence level. For chloride, the limit of detection is about0.014 wt percent Cl for 95 percent confidence, and 0.010,0.008, and 0.005 wt percent for 90, 80, and 60 percent confi-dence levels, respectively.

Besides spot analyses of minerals under manual control, wehave used the electron microprobe to do approximately 1,600point analyses in total on the plus 20-mesh and 20- to 100-mesh screen sizes in the polished plugs under automated con-trol, which allows for 24 h/d use of the instrument. For eachof the two groups of analyses, the time required was about 15h to do the 12 elements listed above.

Electron microprobe results

From the group of elements analyzed by electron micro-probe, it was possible to discover not only the spots with ele-vated concentrations of copper but to ascertain their mineral-ogy as well as that of the minerals hosting them, therebyproviding necessary guidance for subsequent optical exami-nation of inclusions in specific rock-forming minerals. In Fig-ure 3, we show histograms for the spots analyzed by micro-probe on the plus 20-mesh (Fig. 3A) and the 20- to 100-mesh(Fig. 3B) sets of samples. In both cases, over 60 percent of all

the 1,600 spots analyzed contain copper above the limit of de-tection (0.022 wt % or 220 ppm for a 60% confidence level).It should be kept in mind that the electron beam is about 5m in diameter. Hence, the frequency of copper-bearingspots implies occurrence in relatively common minerals,which should be visible using the petrographic microscopewith a high power objective that can discern minerals in ex-cess of 2 or 3 m in size.

In order to interpret the nature of the copper-bearingphase(s), in Figure 4 we plot a series of definitive elements(Cu, Fig. 4A; Cl, Fig. 4B; molar Cu/Cl, Fig. 4C) against K, anelement present at different contents in the several dis-cernible rock-forming minerals of the host rock. Notice inFigure 4C that many of the data points fall near a value of 2.0for molar Cu/Cl, which is representative of ideal stoichiomet-ric atacamite Cu4Cl2(OH)6 and is shown as a horizontal linenear the bottom of Figure 4C. We interpret the data to rep-resent microscopic inclusions of atacamite in 60 percent ofthe analyzed spots. The points with higher Cu/Cl in the dia-gram correspond to analyses where the Cl was very low andin part is due to analytical uncertainty in the Cl data. How-ever, since chrysocolla has also been found in the samples,some of the data plotting along the vertical axis might repre-sent this mineral.

404 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 404

FIG. 3. Histogram of (A) plus 20-mesh and (B) 20- to 100-mesh size sam-ples categorized by wt percent copper in 5-m-diam spots by electron mi-croprobe analysis.

A

B

The identity of the various host minerals containing oc-cluded copper can be inferred from the multielement probedata. In Figure 4A, we show the wt percent Cu as a functionof the wt percent K in four different host minerals, allfeldspars, where copper occurs as inclusions. With increasingK content the minerals are plagioclase, exsolved albite inperthite, orthoclase, and fine-grained K feldspar in thegroundmass of the porphyry host rock. Notice that the copperconcentration ranges up to 2.5 wt percent. The data form dis-cernible groups for each of the inferred host minerals. Thechloride versus K concentrations in wt percent (Fig. 4B) showa pattern similar to that in Figure 4A.

Analogous diagrams for Ca (Fig. 5A-C), Na (Fig. 6A-C), Si(Fig. 7A-C), and Al (Fig. 8A-C) show similar patterns for themicroinclusions of atacamite in gangue mineral rock-formingsilicates, further implicating as copper hosts phenocryst pla-gioclase (Fig. 5), potassium feldspar (Figs. 6 and 7), and mus-covite (Figs. 7 and 8). Figure 9A-C indicates that most mi-croinclusions of atacamite occur within host silicates with Fecontents of 2.2 wt percent or lower, such as Mg-rich biotite inthe potassic alteration zone. The electron microprobe resultsprovided guidance for high-magnification petrographic exam-ination of samples for silicate minerals. These studies confirmthe presence of 1- to 8-m-diameter microinclusions of ata-camite in plagioclase (Fig. 10A) and biotite (Fig. 10B).


0361-0128/98/000/000-00 $6.00 405

FIG. 4. Wt percent (A) Cu, (B) Cl, (C) molar ratio of Cu/Cl, vs. wt per-cent K in plus 20-mesh size samples. Note that a Cu/Cl molar ratio of 2.0 isindicative of atacamite as its ideal formula is Cu4Cl2(OH)6.

FIG. 5. Wt percent (A) Cu, (B) chloride, and (C) molar ratio of Cu/Cl (C)vs. wt percent Ca.

A

B

C

A

B

C

Contribution of microscopic atacamite grains to overall copper

Using composition-volume-density relationships it is possi-ble to determine the contribution of recognized microscopicgrains of atacamite to the overall copper grade of the samplesstudied. The governing relationship between mineral volume,chemical concentration of Cu in the mineral, and its contri-bution to the overall copper grade of the sample follows themethod of Brimhall et al. (1984). The Cu contribution in therock from each mineral is equal to the product of the con-centration of Cu in the mineral (in wt %) times the wt percent

of the mineral divided by 100. To determine the wt percent ofeach mineral, the volumetric modes must be converted to agravimetric basis using mineral density: quartz (2.65 g/cm3),plagioclase (2.64), alkali feldspar (2.57), biotite (2.90), ata-camite (3.77). Modal (volumetric) analysis of the plus 20mesh and 20- to 100-mesh-size fractions were made. We havecomputed the distribution of trace copper using least squaresinversion methods and known compositional variation in thesilicate minerals to find a best fit. The modes and calculatedaverage copper grades for each phase are shown in Table 1 forthe plus 20 mesh samples. The average total contained cop-per (Cutotal) grade of all six samples is 0.38 wt percent. We offeran example of the calculated contribution of Cu in plagioclase

406 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 406

FIG. 6. Wt percent (A) Cu, (B) chloride, and (C) molar ratio of Cu/Cl (C)vs. wt percent Na.

FIG. 7. Wt percent (A) Cu, (B) chloride, and (C) molar ratio of Cu/Cl (C)vs. wt percent Si.

AA

B

C

B

C

to the total copper. The mode of plagioclase is 9.7 percent. Tofind the wt percent plagioclase in the rock, we multiply themodal value of 9.7 percent by the ratio of the density of plagi-olcase to the average density of the minerals present, 2.60 g/cc,giving a value of 9.85 wt percent plagioclase, slightly larger thanits modal volumetric value. The calculated copper content ofplagioclase from the best-fit regression is 0.26 percent. To findthe contribution of plagioclase to the total copper value, wemultiply the wt percent of plagioclase by the wt percent copperit contains and divide by 100 since we express the concentra-tion in percent, yielding 0.026 wt percent copper. Carrying outthis type of calculation for quartz, muscovite, biotite, atacamite,K feldspar, igneous orthoclase, albite, clays, and chrysocolla,

the total microscopic copper sums to 0.29, including atacamite,or 0.13 wt percent, excluding atacamite. Therefore, the calcu-lated microscopically occluded Cu (excluding atacamite) con-sidering all these minerals is (0.13 / 0.38) * 100 = 34 percent oftotal copper of the samples in the plus 20-mesh fraction. Theremaining 66 percent represents the nonmicroscopic Cu, in-cluding atacamite. The amount of Cu by wt percentage con-tained in the various minerals as a percentage of total Cu as mi-croscopic inclusions is quartz (7.7), plagioclase (20), muscovite(0.35), biotite (3.2), hydrothermal K feldspar (42), igneous or-thoclase (23), albite (0.85), and clays (2.2), excluding atacamite.

The analogous results for the 20- to 100-mesh samples areillustrated in Table 2. The average copper grade of all six


0361-0128/98/000/000-00 $6.00 407

FIG. 8. Wt percent (A) Cu, (B) chloride, and (C) molar ratio of Cu/Cl (C)vs. wt percent Al.

FIG. 9. Wt percent (A) Cu, (B) chloride, and (C) molar ratio of Cu/Cl (C)vs. wt percent Fe.

AA

B

C

B

C

samples in this size range is 0.50 wt percent Cutotal, whichmeans the microscopic copper is (0.13/0.50) * 100 = 26 per-cent of the total copper. Thus, by combining results on theplus 20- and the 20- to 100-mesh samples, we conclude thatthe microscopic copper mineral inclusions contained in otherminerals account for about 30 percent of the total copper inthese samples.

How representative these results are for the other size frac-tions besides the plus 20- and the 20- to100-mesh size is eas-ily ascertained. By having screened each of the six samplesinto four separate size fractions and studying the two largerones more amenable to electron microprobing of intact sur-faces, we demonstrated that all samples in each of the fourfractions from coarsest to finest showed the typical increase intotal copper with decreasing size by a factor of two to three.This effect is generally attributed to finer grinding liberatingfracture-controlled mineralization. However, the ratio of acidsoluble copper to total copper (a range of 85100%) re-mained roughly constant (within 5%) over all size ranges.Hence, we are certain that in all size ranges, these microin-clusions of atacamite do in fact represent most of the insolu-ble copper fraction of the samples poor in clays and most of it(80 wt %) is in rock-forming feldspars.

Copper-Bearing Clay Ore

Sample preparation

The eight samples collected for this part of the study wereexamined under a stereomicroscope, and representative por-tions of different clay domains of various colors were hand-picked for mineralogical and chemical analysis. The sampledclay domains include veinlets and fracture coatings, shapelessmicrocrystalline aggregates and patches of soaplike appear-ance, and pseudomorphs after feldspar judging from theirhabit. Microfractured inhomogeneous parts with atacamiteand chrysocolla visible under the stereomicroscope wereavoided. The colors of the clay minerals vary in differentshades of green and yellow, to almost white. Before analyzingthe separated clays with X-ray diffraction, a few representa-tive grains were selected from each separate for microchemi-cal analysis and crushed to smaller particles by simple pres-sure (no grinding).

Analytical methods

Twenty-two X-ray diffractograms (Cu radiation; scans from2 to 65 2) with count data collected at intervals of 0.02 2for a time of 1.25 s were obtained of seemingly homogeneous

408 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 408

TABLE 1. Modes and Calculated Average Copper Grades for Each Phase (plus 20 mesh)

% Cu in ContributionVol mineral Bulk Wt to total

Mineral percent regression density percent Cu (%)

Quartz 38 0.03 2.65 38.73 0.012Plagioclase 9.7 0.26 2.64 9.85 0.026Muscovite 0.6 0.07 2.8 0.65 0.00046Biotite 2.2 0.17 2.9 2.45 0.0042Atacamite 0.2 57 3.77 0.29 0.16K feldspar 29 0.19 2.57 28.66 0.054Igneous orthoclase 13 0.24 2.57 12.85 0.031

Albite 1.6 0.07 2.62 1.61 0.0011Chrysocolla 5.6 0.06 2.2 4.74 0.0028_____ _______

2.6 0.13

TABLE 2. Modes and Calculated Average Copper Grades for Each Phase (20100 mesh)

% Cu in ContributionVol mineral Bulk Wt to total

Mineral percent regression density percent Cu (%)

Quartz 39 0.03 2.65 39.75 0.012Plagioclase 11 0.32 2.64 11.17 0.036Muscovite 2.1 0.03 2.8 2.27 0.00068Biotite 1.8 0.67 2.9 2.0 0.013K feldspar 25 0.13 2.57 24.71 0.032Igneous orthoclase 11 0.22 2.57 10.87 0.0224

Albite 5.7 0.11 2.62 5.74 0.0063Chrysocolla 4.9 0.19 2.2 4.14 0.0079_____ _______

2.60 0.13

FIG. 10. Photomicrograph of disseminated atacamite microinclusions in (A) plagioclase and (B) biotite shown at samescale.

A B

0 10 20 30 40 50m

clay mineral domains without visible grains of other phasesunder the stereomicroscope. Sixteen of these samples werererun from 3 to 16 after ethylene glycol solvation in orderto check the swelling character of the clays. In addition, 16determinations of green and blue Cu minerals, clay with relictfeldspar, and illite were made. The X-ray diffraction data forthe clays were interpreted based on information in Thorez(1976), Bailey (1991), and the International Centre for Dif-fraction Data (1993). A total number of 183 chemical analy-ses were made with energy dispersive spectroscopy, using ascanning electron microscope in different domains, 141 ofthem in clay minerals, 11 in Cu chlorides and silicates, and 14in relict feldspar, illite, biotite, and apatite. Fifteen analyseswere discarded because the analyzed grain contained morethan one phase. This method gives less accurate data thananalysis with electron microprobe, especially with regard tolight elements. However, Cu is sufficiently well resolved thatthe results serve a useful purpose of internal comparisons.Both the X-ray diffractometry and energy dispersive spec-trometry were made at the Swedish Museum of Natural His-tory, Stockholm.

Copper-Bearing Smectites

Smectite composition

Chemically, the analyzed clay minerals from RadomiroTomic are rather uniform with respect to Si, Mg, Ca, Na, andK but highly variable in Al, Fe, and Cu (Table 3). They arecharacterized by high Si contents (SiO2 = 4857 wt %), mod-erate to very high Al (Al2O3 = 1036 wt %), low to moderateFe (Fe, for comparative purposes given as FeO = 112 wt %;however, most of the Fe is probably trivalent), very low tomoderate Mg, Ca, Na, and K (MgO = 0.02.6, CaO = 0.02.3,Na2O = 0.02.5, and K2O = 0.01.4 wt %). Copper variesfrom 0.1 to 23.0 wt percent (Fig. 11), and Cl is either absent(samples DDH-3758, RT-13014, and RT-13923) or present insmall amounts, usually less than 0.4 wt percent; in one sam-ple up to 0.7 wt percent). Some differences in chemical com-position can be related to site. Generally, the veinlets arericher in Cu than patches in the rock, which in their turn tendto have higher Cu contents than pseudomorphs after feldspar.In addition, domains of stronger color are usually richer in Cuthan those of paler shades. Veinlets are also richer in Fe andpoorer in Al and Mg than the pseudomorphs after feldspar.

In a plot of interlayer cation total (Ca + Na + K) versus Si,the high Si values (between 8.5 and 10 based on 28 oxygens)of the clay minerals identify them as smectites (Fig. 12).Chlorites normally have values less than 6.25, and mixed-layerinterstratified smectite-chlorite phases plot between 6.25 and8 (cf. Bettison and Schiffman, 1988; Schmidt and Robinson,1997, and references therein). It cannot be excluded thatanalyses with Si values in the 8 to 9 range represent inter-stratified phases. These analyses correspond to Cu-poorsmectites replacing feldspar. However, all the clay minerals insamples DDH-3758 and RT-13007 are pure smectites, eventhose in feldspar sites, like the large majority of Cu-bearingsmectites in the other samples (Fig. 12). Other compositionalfeatures consistent with a smectitic nature are the moderatelyhigh sums for the interlayer cations (Ca + Na + K) and therather high Ca values for the Cu-bearing smectites (Fig. 12,


0361-0128/98/000/000-00 $6.00 409

FIG. 11. Distribution of copper in different domains of smectite in sam-ples from the oxidation zone of Radomiro Tomic (wt % Cu based on sum ofoxides recalculated to a total of 88 wt %). Each square represents one analy-sis. The letters V, R, or F stand for smectite in veinlets (and coated planes),rock (= patches in rock), and pseudomorphs after feldspar, respectively, fol-lowed by a second letter indicating color: a = apple-green, d = dull olive-green, g = unspecified yellowish-green to green, p = pistachio-green, w =waxy olive-green, and y = yellow shades.

410 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 410

TAB

LE

3. R

epre

sent

ativ

e C

hem

ical

Ana

lyse

s (E

nerg

y D

ispe

rsiv

e Sp

ectr

osco

py)

of S

mec

tites

from

the

Oxi

datio

n Zo

ne o

f Rad

omir

o To

mic

Sam

ple

DD

H-3

758;

151

.9 m

RT-

2915

-AR

T-29

15-B

Site

Vg2

Vg1

Vg3

FV

yR

p

Ana

lysi

s24

2933

Ave

rage

Ran

ge4

1316

Ave

rage

Ran

ge61

Ave

rage

Ran

ge52

Ave

rage

Ran

ge7

Ran

ge7

1118

Ave

rage

Ran

ge(1

9)(1

9)(2

1)(2

1)(5

)(5

)(6

)(6

)(2

)(2

1)(2

1)

SiO

2(w

t %)

53.6

651

.04

53.1

053

.44

49.5

56.

454

.19

54.6

154

.04

54.5

853

.45

5.6

51.2

252

.50

51.2

53.

755

.70

55.5

755

.25

5.8

53.7

951

.35

3.8

51.7

753

.47

53.3

153

.78

50.8

56.

4A

l 2O3

19.3

114

.75

17.6

317

.79

12.5

20.

113

.93

17.6

116

.68

16.9

213

.91

8.1

22.3

222

.66

22.1

23.

522

.20

22.8

821

.62

3.8

17.1

817

.22

3.1

28.8

422

.54

19.9

223

.04

19.3

29.

6F

eO8.

054.

686.

316.

733.

411

.97.

949.

108.

608.

797.

810

.78.

797.

205.

38.

85.

024.

493.

65.

410

.39

8.1

10.4

3.76

6.43

7.50

5.43

3.8

7.7

MgO

1.72

0.80

1.00

1.14

0.7

1.7

1.28

1.34

1.34

1.27

0.8

1.7

1.62

1.56

1.4

1.7

1.74

1.68

1.5

2.0

0.00

0.0

0.8

0.31

1.03

1.38

1.28

0.3

1.8

CaO

1.03

0.94

0.97

1.03

0.8

1.3

1.25

1.02

1.17

1.13

0.9

1.3

1.13

1.19

1.0

1.4

1.14

1.14

1.1

1.2

0.15

0.0

0.2

0.78

1.49

1.60

1.35

0.7

2.3

Na 2

O0.

080.

810.

370.

400.

01.

00.

660.

340.

530.

300.

00.

70.

230.

110.

00.

30.

180.

210.

00.

72.

482.

02.

51.

080.

810.

901.

140.

32.

3K

2O0.

750.

720.

730.

590.

31.

20.

200.

530.

280.

390.

11.

10.

120.

280.

10.

70.

280.

220.

20.

30.

270.

30.

50.

080.

350.

450.

210.

10.

7C

uO3.

4014

.27

7.89

6.87

3.2

14.3

8.54

3.47

5.36

4.62

3.2

8.5

2.57

2.50

2.3

2.7

1.73

1.81

1.7

2.0

3.75

2.3

3.8

1.39

1.88

2.95

1.76

1.3

3.0

C1

0.00

0.00

0.00

0.00

0.0

0.0

0.00

0.07

0.00

0.02

0.0

0.1

0.00

0.00

0.0

0.0

0.00

0.00

0.0

0.0

0.44

0.3

0.4

ndnd

0.59

0.36

0.0

0.7

Sum

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

0C

u2.

7211

.40

6.31

5.49

2.5

11.4

6.82

2.77

4.28

3.69

2.5

6.8

2.05

2.00

1.8

2.2

1.38

1.45

1.3

1.6

3.00

1.8

3.0

1.11

1.50

2.35

1.41

1.0

2.4

At.

prop

ortio

n ba

sed

on 2

2 ox

ygen

sSi

7.62

7.68

7.69

7.70

7.94

7.78

7.78

7.82

7.26

7.36

7.64

7.60

7.77

7.07

7.46

7.55

7.45

Al(I

V)

0.38

0.32

0.31

0.33

0.06

0.22

0.22

0.18

0.74

0.64

0.36

0.40

0.23

0.93

0.54

0.45

0.55

(I

V)

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

Al(V

I)2.

852.

292.

702.

712.

352.

742.

612.

683.

003.

113.

233.

292.

703.

713.

162.

883.

21F

e0.

960.

590.

760.

810.

971.

081.

041.

051.

040.

850.

580.

511.

260.

430.

750.

890.

63M

g0.

360.

180.

220.

240.

280.

280.

290.

270.

340.

330.

360.

340.

000.

060.

210.

290.

26C

u0.

371.

620.

860.

750.

950.

370.

580.

500.

280.

270.

180.

190.

410.

140.

200.

320.

19

(VI)

4.53

4.68

4.54

4.52

4.55

4.49

4.52

4.50

4.66

4.55

4.35

4.34

4.37

4.35

4.33

4.38

4.29

Ca

0.16

0.15

0.15

0.16

0.20

0.16

0.18

0.17

0.17

0.18

0.17

0.17

0.02

0.11

0.22

0.24

0.20

Na

0.02

0.23

0.10

0.11

0.19

0.09

0.15

0.08

0.06

0.03

0.05

0.06

0.69

0.29

0.22

0.25

0.31

K0.

140.

140.

130.

110.

040.

100.

050.

070.

020.

050.

050.

040.

050.

010.

060.

080.

04

At.

prop

ortio

n ba

sed

on 2

8 ox

ygen

sSi

9.69

9.77

9.78

9.80

10.1

19.

919.

909.

959.

249.

379.

739.

689.

899.

009.

499.

619.

49A

l4.

113.

333.

833.

843.

063.

773.

603.

644.

754.

774.

574.

703.

735.

914.

724.

234.

79F

e1.

220.

750.

971.

031.

241.

381.

321.

341.

331.

080.

730.

651.

600.

550.

951.

130.

80M

g0.

460.

230.

280.

310.

360.

360.

370.

340.

440.

420.

450.

440.

000.

080.

270.

370.

34C

u0.

462.

061.

100.

961.

200.

480.

740.

640.

350.

340.

230.

240.

520.

180.

250.

400.

24C

a0.

200.

190.

190.

200.

250.

200.

230.

220.

220.

230.

210.

210.

030.

140.

280.

310.

26N

a0.

030.

300.

130.

140.

240.

120.

190.

110.

080.

040.

060.

070.

880.

360.

280.

310.

39K

0.17

0.18

0.17

0.14

0.05

0.12

0.07

0.09

0.03

0.06

0.06

0.05

0.06

0.02

0.08

0.10

0.05

in

ter-

laye

r0.

400.

670.

490.

490.

530.

440.

490.

420.

330.

330.

340.

330.

980.

530.

640.

730.

70

noni

nter

-la

yer

15.9

516

.14

15.9

615

.94

15.9

715

.89

15.9

415

.91

16.1

115

.97

15.7

115

.70

15.7

415

.71

15.6

915

.75

15.6

6

Not

es: T

he le

tter

s V,

R, a

nd F

sta

nd fo

r sm

ectit

e in

vei

nlet

s (a

nd c

oate

d pl

anes

), ro

ck (

= pa

tche

s in

roc

k), a

nd p

seud

omor

phs

afte

r fe

ldsp

ar, r

espe

ctiv

ely,

follo

wed

by

a se

cond

lett

er in

dica

ting

colo

r: g

= u

nspe

ci-

fied

yello

wis

h-gr

een

to g

reen

, a =

app

le-g

reen

, p =

pis

tach

io-g

reen

, d =

dul

l oliv

e-gr

een,

w =

wax

y ol

ive-

gree

n, a

nd y

= y

ello

w s

hade

sN

umbe

r of

ana

lyse

s av

erag

ed in

par

enth

eses

; sum

of o

xide

s re

calc

ulat

ed to

a to

tal o

f 88

wt %

(ex

clud

ing

Cl);

Al(I

V)

and

Al(V

I) c

orre

spon

d to

Al i

n te

trah

edra

l and

oct

ahed

ral p

ositi

on, r

espe

ctiv

ely;

(V

I) v

alue

sin

clud

e C

u;

inte

rlay

er =

sum

of i

nter

laye

r ca

tions

(C

a +

Na

+ K

) an

d

noni

nter

laye

r =

sum

of n

onin

terl

ayer

cat

ions

, inc

ludi

ng C

u (S

i + A

l + F

e +

Mg

+ C

u)


0361-0128/98/000/000-00 $6.00 411

TAB

LE

3.(C

ont.)

Sam

ple

RT-

2915

-BR

T-29

15-D

RT-

1300

7

Site

Ra

Rd

Rw

Ry

Rg

Ry

Va

Ana

lysi

s60

61A

vera

geR

ange

3238

Ave

rage

Ran

ge45

Ave

rage

Ran

ge57

Ave

rage

Ran

ge6

Ran

ge2

89

Ave

rage

Ran

ge(7

)(7

)(1

0)(1

0)(6

)(6

)(6

)(6

)(2

)(5

)(5

)

SiO

2(w

t %)

55.0

155

.05

55.9

455

.05

7.1

52.3

751

.58

52.2

050

.85

3.4

56.6

856

.40

54.8

57.

155

.03

54.9

654

.25

5.9

50.5

750

.65

3.3

51.5

956

.70

53.0

955

.14

53.1

56.

7A

1 2O

320

.95

22.1

519

.67

17.1

22.

227

.21

29.8

928

.88

26.9

31.

420

.57

21.6

720

.32

4.2

25.4

324

.90

23.5

25.

929

.39

27.4

29.

429

.58

19.2

816

.92

19.5

216

.92

2.2

FeO

5.30

4.87

5.83

4.9

7.4

4.90

3.33

3.72

2.7

4.9

4.20

3.84

3.0

4.8

1.26

2.13

1.3

4.0

4.99

2.9

5.0

1.92

5.50

9.88

6.38

4.7

9.9

Mgo

1.76

1.59

1.85

1.6

2.1

0.30

0.64

0.50

0.0

1.1

2.56

2.31

1.9

2.6

1.65

1.56

1.0

1.9

0.32

0.3

0.3

1.43

2.30

1.52

1.75

1.4

2.3

CaO

1.28

1.22

1.33

1.2

1.4

0.87

0.72

0.78

0.5

1.0

1.32

1.21

0.9

1.4

1.41

1.25

0.9

1.4

0.86

0.9

1.4

1.11

0.85

1.33

1.16

0.9

1.4

Na 2

O1.

631.

641.

751.

42.

50.

330.

250.

330.

00.

91.

551.

461.

31.

61.

981.

681.

32.

00.

140.

10.

70.

901.

971.

441.

711.

42.

0K

2O0.

130.

120.

140.

10.

30.

230.

230.

170.

10.

30.

120.

180.

10.

40.

700.

970.

51.

40.

120.

10.

40.

450.

210.

120.

130.

10.

2C

uO1.

941.

361.

491.

31.

91.

791.

361.

431.

11.

80.

990.

920.

71.

10.

530.

540.

10.

91.

611.

61.

71.

011.

183.

722.

211.

23.

7C

l0.

500.

420.

430.

20.

60.

150.

060.

160.

00.

30.

300.

380.

20.

60.

570.

400.

30.

60.

030.

00.

40.

310.

100.

480.

300.

10.

5Su

m88

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

Cu

1.55

1.09

1.19

1.1

1.6

1.43

1.09

1.14

0.9

1.5

0.79

0.74

0.6

0.9

0.43

0.43

0.1

0.7

1.29

1.3

1.3

0.81

0.94

2.97

1.77

0.9

3.0

At.

prop

ortio

n ba

sed

on 2

2 ox

ygen

sSi

7.63

7.58

7.76

7.19

7.01

7.10

7.75

7.69

7.53

7.46

6.95

6.99

7.83

7.67

7.71

Al(I

V)

0.37

0.42

0.24

0.81

0.99

0.90

0.25

0.31

0.56

0.54

1.05

1.01

0.17

0.33

0.29

(I

V)

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

Al(V

I)3.

053.

182.

973.

603.

793.

743.

073.

183.

493.

453.

713.

722.

962.

552.

92F

e0.

610.

560.

680.

560.

380.

420.

480.

440.

140.

240.

570.

220.

641.

190.

75M

g0.

360.

330.

380.

060.

130.

100.

520.

470.

330.

320.

070.

290.

470.

330.

36C

u0.

200.

140.

160.

190.

140.

150.

100.

090.

050.

060.

170.

100.

120.

410.

24

(VI)

4.24

4.21

4.19

4.41

4.44

4.41

4.18

4.18

4.01

4.06

4.52

4.33

4.20

4.47

4.27

Ca

0.19

0.18

0.20

0.13

0.10

0.11

0.19

0.18

0.20

0.18

0.13

0.16

0.13

0.21

0.17

Na

0.44

0.44

0.47

0.09

0.07

0.09

0.41

0.39

0.52

0.44

0.04

0.24

0.53

0.40

0.46

K0.

020.

020.

030.

040.

040.

030.

020.

030.

120.

170.

020.

080.

040.

020.

02

At.

prop

ortio

n ba

sed

on 2

8 ox

ygen

sSi

9.71

9.65

9.87

9.16

8.92

9.04

9.87

9.79

9.46

9.50

8.84

8.90

9.96

9.76

9.81

Al

4.36

4.58

4.09

5.61

6.09

5.90

4.22

4.43

5.16

5.07

6.06

6.02

3.99

3.67

4.09

Fe

0.78

0.71

0.86

0.72

0.48

0.54

0.61

0.56

0.18

0.31

0.73

0.28

0.81

1.52

0.95

Mg

0.46

0.42

0.49

0.08

0.16

0.13

0.66

0.60

0.42

0.40

0.08

0.37

0.60

0.42

0.46

Cu

0.26

0.18

0.20

0.24

0.18

0.19

0.13

0.12

0.07

0.07

0.21

0.13

0.16

0.52

0.30

Ca

0.24

0.23

0.25

0.16

0.13

0.15

0.25

0.23

0.26

0.23

0.16

0.20

0.16

0.26

0.22

Na

0.56

0.56

0.60

0.11

0.08

0.11

0.52

0.49

0.66

0.56

0.05

0.30

0.67

0.51

0.59

K0.

030.

030.

030.

050.

050.

040.

030.

040.

150.

210.

030.

100.

050.

030.

03

in

ter-

laye

r0.

830.

810.

880.

330.

270.

290.

800.

761.

071.

010.

240.

610.

880.

800.

84

noni

nter

-la

yer

15.5

715

.54

15.5

115

.80

15.8

415

.79

15.5

015

.50

15.2

915

.35

15.9

315

.69

15.5

215

.88

15.6

2

412 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 412

TAB

LE

3.(C

ont.)

Sam

ple

RT-

1300

7R

T-13

014

RT-

1392

3

Site

Vy

RF

Vg

Vy

FF

Ana

lysi

s4

5A

vera

geR

ange

16R

ange

14R

ange

915

1617

Ave

rage

Ran

ge2

3A

vera

geR

ange

7R

ange

7A

vera

geR

ange

(5)

(5)

(2)

(3)

(10)

(10)

(4)

(4)

(2)

(5)

(5)

SiO

2(w

t %)

53.5

556

.01

54.7

453

.65

6.0

56.0

456

.05

6.5

54.5

053

.85

4.5

43.6

648

.98

48.8

549

.98

48.6

143

.75

0.5

49.4

352

.32

51.0

549

.45

2.3

51.0

048

.95

1.0

52.0

450

.34

47.9

52.

0A

1 2O

318

.65

20.9

719

.86

18.7

21.

024

.10

24.1

26.

725

.19

25.2

25.

69.

5517

.65

12.4

124

.77

18.7

49.

624

.815

.91

28.9

724

.96

15.9

29.

032

.23

32.2

33.

524

.57

30.5

324

.63

5.5

FeO

7.85

4.56

6.27

4.7

7.9

2.17

0.8

2.2

1.44

1.4

1.7

3.03

7.57

4.45

8.77

7.41

3.0

11.3

3.36

4.49

4.88

3.4

6.0

2.28

1.9

2.3

5.86

3.92

2.2

5.9

MgO

1.68

2.22

2.06

1.7

2.3

1.79

0.9

1.8

1.78

1.8

1.9

0.61

0.46

0.20

0.66

0.62

0.0

1.5

0.72

0.38

0.27

0.0

0.7

0.99

1.0

1.0

1.67

0.69

0.0

1.7

CaO

1.48

1.13

1.31

1.0

1.5

1.16

0.8

1.2

1.59

1.3

1.6

0.98

0.70

1.41

0.46

0.78

0.4

1.4

0.75

0.30

0.48

0.3

0.8

0.50

0.4

0.5

0.58

0.43

0.3

0.6

Na 2

O2.

112.

001.

921.

72.

11.

411.

41.

51.

971.

92.

31.

230.

661.

490.

000.

680.

01.

51.

310.

000.

330.

01.

30.

190.

20.

30.

730.

200.

00.

7K

2O0.

040.

100.

090.

00.

20.

150.

20.

20.

360.

40.

60.

190.

220.

420.

110.

260.

10.

50.

210.

080.

180.

10.

20.

180.

20.

21.

160.

670.

21.

2C

uO2.

641.

011.

741.

02.

61.

180.

61.

21.

161.

01.

328

.75

11.7

618

.78

3.25

10.8

93.

328

.816

.32

1.46

5.85

1.5

16.3

0.64

0.6

0.8

1.41

1.21

0.9

1.8

Cl

0.38

0.37

0.40

0.3

0.6

0.11

0.1

0.2

0.38

0.3

0.5

0.06

0.00

0.00

0.00

0.01

0.0

0.1

0.00

0.00

0.00

0.0

0.0

0.00

0.0

000.

000.

000.

00.

0Su

m88

.00

88.0

088

0088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

88.0

088

.00

Cu

2.11

0.80

1.39

0.8

2.1

0.95

0.4

1.0

0.93

0.8

1.1

22.9

79.

3915

.00

2.59

8.70

2.6

23.0

13.0

41.

174.

681.

113

.00.

510.

50.

61.

130.

970.

71.

5

At.

prop

ortio

n ba

sed

on 2

2 ox

ygen

sSi

7.61

7.69

7.64

7.57

7.40

7.29

7.35

7.62

7.09

7.26

7.48

7.12

7.20

6.85

7.24

6.88

Al(I

V)

0.39

0.31

0.36

0.43

0.60

0.71

0.65

0.38

0.91

0.74

0.52

0.88

0.80

1.15

0.76

1.12

(I

V)

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

8.00

Al(V

I)2.

733.

092.

913.

413.

431.

172.

471.

903.

232.

532.

323.

773.

323.

963.

283.

80F

e0.

930.

520.

730.

250.

160.

420.

950.

581.

040.

920.

420.

510.

570.

260.

680.

45M

g0.

360.

460.

430.

360.

360.

150.

100.

050.

140.

140.

160.

080.

060.

200.

350.

14C

u0.

280.

100.

180.

120.

123.

631.

332.

210.

351.

271.

870.

150.

650.

070.

150.

13

(VI)

4.31

4.17

4.26

4.14

4.07

5.37

4.86

4.74

4.76

4.85

4.77

4.51

4.60

4.48

4.45

4.51

Ca

0.23

0.17

0.20

0.17

0.23

0.18

0.11

0.23

0.07

0.13

0.12

0.04

0.07

0.07

0.09

0.06

Na

0.58

0.53

0.52

0.37

0.52

0.40

0.19

0.45

0.00

0.20

0.38

0.00

0.10

0.05

0.20

0.05

K0.

010.

020.

020.

030.

060.

040.

040.

080.

020.

050.

040.

010.

030.

030.

210.

12

At.

prop

ortio

n ba

sed

on 2

8 ox

ygen

sSi

9.69

9.79

9.73

9.64

9.41

9.28

9.36

9.70

9.02

9.24

9.52

9.06

9.17

8.72

9.22

8.76

Al

3.98

4.32

4.16

4.89

5.13

2.39

3.97

2.90

5.27

4.16

3.61

5.91

5.24

6.50

5.13

6.25

Fe

1.19

0.67

0.93

0.31

0.21

0.54

1.21

0.74

1.32

1.17

0.54

0.65

0.73

0.33

0.87

0.57

Mg

0.45

0.58

0.55

0.46

0.46

0.19

0.13

0.06

0.18

0.17

0.21

0.10

0.08

0.25

0.44

0.18

Cu

0.36

0.13

0.23

0.15

0.15

4.62

1.70

2.82

0.44

1.62

2.37

0.19

0.83

0.08

0.19

0.16

Ca

0.29

0.21

0.25

0.21

0.29

0.22

0.14

0.30

0.09

0.16

0.15

0.06

0.09

0.09

0.11

0.08

Na

0.74

0.68

0.66

0.47

0.66

0.51

0.24

0.57

0.00

0.26

0.49

0.00

0.12

0.06

0.25

0.07

K0.

010.

020.

020.

030.

080.

050.

050.

110.

020.

060.

050.

020.

040.

040.

260.

15

inte

r-la

yer

1.04

0.91

0.93

0.72

1.03

0.78

0.44

0.98

0.11

0.48

0.70

0.07

0.26

0.19

0.62

0.30

no

nint

er-

laye

r15

.66

15.4

915

.60

15.4

515

.36

17.0

216

.37

16.2

116

.24

16.3

616

.25

15.9

216

.04

15.8

815

.85

15.9

2

Table 3; cf. Schmidt and Robinson, 1997). Also, in this casethe exceptions are found among phases replacing feldspar.

A plot of total noninterlayer cations (Si + Al + Fe + Mg)against total Al cations (Fig. 13; Schiffman and Fridleifsson,1991) shows that the smectites rich in Al occupy a field char-acteristic of dioctahedral smectites, i.e., the montmorillonitesubgroup (Fig. 13A). Smectites containing less Al define a di-vergent trend toward anomalous compositions. Two sampleswith Cu-rich smectites, DDH-3758 and RT-13014, are largelyresponsible for the divergent trend (Fig. 13C and E), whichdisappears if Cu is treated as a noninterlayer cation. Now, thesmectites with less Al define a new trendtoward the field oftrioctahedral smectites, i.e., the saponite subgroup (Fig. 13B,D, and F).

The sum of cations in octahedral position (VI) also givesstructural information. Theoretically, the value should be 4for dioctahedral and 6 for trioctahedral phases (based on 22oxygens). Table 3 shows that the analyzed smectites have(VI) values in the range 4 to 4.7 if Cu is included (up to 4.9or more for sample RT-13014). If Cu would not be taken intoaccount, then too low values (down to

similar to those of dioctahedral smectites (montmorillonites)given in the literature (International Centre for DiffractionData, 1993). The (060) reflection for the analyzed smectiteshas values in the range 1.485 to 1.498 (Fig. 15), which iswell below the limit of 1.52 for the dioctahedral subgroup(Thorez, 1976). In spite of being dioctahedral, a trend towarda trioctahedral character can be discernedthe d-spacing forthe (060) reflection increases with the Cu content (Fig. 15).This trend is defined by the smectites with considerable tohigh contents of Cu that occur in veinlets, and it coincideswith the conclusions based on chemistry.

Two distinct peaks of considerable intensity, at ca. 7.3 (12.1 2) and 3.6 (24.7 2; Table 4, Fig. 14), pose a prob-lem. They correspond to the (002) and (004) reflections, re-spectively, and are present in all the samples except DDH-3758 and RT-13007. The existence of these peaks and a ratherhigh value of the ratio between the intensities of the (002)and (001) reflections (= 7/14 ; Table 4, Fig. 14) might betaken as evidence of mixed-layer smectite-chlorite phases (cf.Schmidt and Robinson, 1997), in contradiction to the conclu-sions based on chemistry. The symmetry of the (001) reflec-tion in the glycolated runs (Fig. 14) also argues against inter-stratification with chlorite (Thorez, 1976). The 7.3 peakcannot be caused by a discrete chlorite phase or presence ofkaolinite. Presence of chlorite and/or kaolinite would have re-sulted in more than one population among the chemically an-alyzed grains of a clay domain and not give relatively homo-geneous compositions of high Si content. The problematic 7.3and 3.6 peaks are unrelated to the Cu (and Fe) content ofthe smectites; they appear in smectites rich as well as poor inCu (Table 4, Fig. 14). The effect of Cu substitution on theXRD patterns of smectites is unknown. It should be notedthat the Cu smectite yakhontovite (CuO = 15.026.0 wt %)has a 7.3 (12.1 2) peak without being classified as amixed-layer phase (Postnikova et al., 1986; Jambor andPuziewicz, 1991; International Centre for Diffraction Data,1993).

Copper in the crystal structure

The main evidence indicating that copper is structurallybound in the smectites of Radomiro Tomic is (1) the positivecorrelation between Cu and structurally dependent parame-ters, such as the d-spacing of the (060) reflection (Fig. 15); (2)the change from anomalous compositions to a normal trend ifCu is treated as another cation in the plot against Al (Fig. 13);and (3) the change from a poor to an almost perfect negativecorrelation between Fe + Mg alone on the one hand, and Fe+ Mg with Cu added on the other, in plots against Al(VI) (Fig.16). This correlation is valid for smectites with considerableto high contents of Cu, as can be seen in a related plot of Cuagainst the cations that normally occupy octahedral positionsin smectites (Al(VI), Fe, and Mg; Fig. 17AC). Observations(2) and (3) suggest that Cu occupies octahedral sites in thestructure, which is consistent with experimental work on tran-sition metals in smectites and chlorites (Mosser et al., 1990,1992; Bailey, 1991; Gven, 1991; Decarreau et al., 1992). Atlow concentrations of Cu (


0361-0128/98/000/000-00 $6.00 415

FIG. 14. Representative X-ray diffractograms of smectites from Radomiro Tomic. The upper eight patterns illustrate air-dried preparations and the two patterns at bottom, raised 50 counts, exemplify the changes after ethylene glycol solvation(cf. Table 4).

structural (Fig. 17DF). These smectites have (VI) valuesup to ca. 0.5 atoms per formula unit above the theoreticalvalue of 4 for dioctahedral smectites.

ConclusionsThis study demonstrates that the acid-insoluble fraction of

atacamite zone copper at Radomiro Tomic, as about 30 per-cent of the total Cu in feldspar-stable portions of the orewhere clays are uncommon, cannot be attributed to the pres-ence of copper sulfides as was earlier suspected but rather isdue to microinclusions of atacamite with diameters on the orderof 1 to 8 m contained in rock-forming feldspars and othersilicate minerals. The soluble copper fraction that is over 70

416 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 416

FIG. 15. Plot of wt percent Cu vs. d-spacing of the (060) reflection forsmectites in veinlets (Cu-rich population) and rock sites (Cu-poor popula-tion) in samples from Radomiro Tomic.

FIG. 16. Plots of Fe + Mg (A, C, and E) and Fe + Mg + Cu (B, D, and F)vs. Al(VI) for smectites from Radomiro Tomic (at. proportions based on 22oxygens; Al(VI) = Al in octahedral position). See Figure 11 for smectiteabbreviations.

FIG. 17. Plots of atoms per formula unit Cu vs. Al(VI) + Fe + Mg forsmectites from Radomiro Tomic (at. proportions based on 22 oxygens; Al(VI)= Al in octahedral position). A-C illustrate the trend for analyses with con-siderable to high Cu contents, whereas D-F show the lack of correlation forsmectites with less Cu. See Figure 11 for smectite abbreviations.

wt percent of the total Cu occurs generally as atacamite incracks, which upon crushing is exposed along surfaces of therock fragments. The disseminated microinclusions of ata-camite are rarely exposed by crushing and remain inside oftheir host silicate minerals.

In contrast, in copper clay ores, in zones with weak,feldspar-destructive argillic alteration caused by supergenesolutions, the occluded copper is structurally bound in octa-hedral sites of well-crystallized smectitic clays. The two dif-ferent acid-insoluble components are products of contrastingmicrochemical environments in a reactive potassium silicatealteration gangue. In both cases, the hydrothermal alterationwas sufficiently weak to leave intact rock-mineral buffers ca-pable of neutralizing acid and reducing fluids. This processcontrolled the precipitation of copper on all scalesthemacrofield zonal scale (Cuadra and Rojas, 2001), the mi-croscale of individual crystals in cracks, and the atomic scaleof octahedral sites in the clay structure. It is very unlikelythat the considerable to high copper contents of manysmectites analyzed here are the consequence of includedparticles and microveinlets of atacamite. The general lack ofcorrelation between Cu and Cl (Fig. 18A-B) for sampleswith greater than 3 percent Cu argues against it. However,for samples with less than 3 percent Cu, the weak correlationof Cu and Cl suggests a possible exisitence of atacamite

microinclusions. The most Cu rich smectites, in samplesDDH-3758 and RT-13014, are Cl free (Cl values less than0.1 wt % are not significant; Table 3). Chlorine-bearingsmectites with Zn instead of Cu in hydrothermally altereddacitic rocks were reported by Pons et al. (1989; Zn was notdetected in the Radomiro Tomic clays). It is also unlikely thatthe considerable to high Cu contents are due to includedchrysocolla because this would not explain the weak negativecorrelation between CuO and SiO2 (Fig. 18D). Microinclu-sions of atacamite might, nevertheless, account for part ofthe Cu in smectites with relatively low Cu contents formingpseudomorphs after feldspar (Figs. 17D-F, 18C), as shownby the lack of correlation between Cu and the sum of Al(VI),Fe, and Mg.

The Radomiro Tomic smectites resemble copper-bearingclay minerals described from two other deposits (Table 5)amegascopically similar sample from the oxidation zone of thePotrerillos copper porphyry, 460 km south of RadomiroTomic (CC-1; Levi, 1997) and the Cu smectite yakhontovitereported from an oxidized Cu-Sn deposit in Russia (Post-nikova et al., 1986; see summary in Jambor and Puziewicz,1991). The smectites of the three deposits have yellowish-green colors in common and copper seemingly occupies octa-hedral sites (Fig. 19). A chemical comparison (Table 5, Fig.19) indicates that structurally bound Cu can occur in differ-ent species: (1) di- to trioctahedral smectites at RadomiroTomiclow Cu grains have a typical dioctahedral chemistryand (060) reflection (montmorillonites), whereas Cu-richgrains show a trend toward a trioctahedral, saponitic chem-istry; (2) saponites and mixed-layer smectite-chlorite phasesat Potrerillos with values of ca. 1.53 for the (060) reflection;(3) the Cu smectite yakhontovite, which is classified as dioc-tahedral in the International Centre for Diffraction Data(1993) but described as dioctahedral with a certain trioctahe-dral character by Postnikova et al. (1986); the dioctahedralvalue for the (060) reflection (1.518 ) is contradicted by atrioctahedral chemistry (Fig. 19C).

Yellowish-green copper-bearing clay minerals have beenfound in the oxidation zone of several porphyry copper de-posits in northern Chile (E. Tidy, unpub. information). In ad-dition, smectites with structurally bound transition metalssuch as Ni, Co, Zn, Cu, and Mn are known to occur in the su-pergene part of some ore deposits (Paquet et al., 1987; Gven,1991) and in the vicinity of smelters (Rybicka and Jedrze-jczyk, 1995; see Cuadra and Rojas, 2001, for one viewpoint).

The microinclusions of atacamite in rock-forming mineralsand copper-bearing smectites documented in this study owetheir existence to the relatively small amount of feldspar-de-structive hydrothermal alteration that has affected RadomiroTomic. In contrast, many other South American porphyrycopper deposits have experienced strong, late-stage hydro-lytic (sericitic) alteration and, hence, offer less reactivity toapplied acid solutions. The conditions of formation of the oc-cluded copper phases is thought to be low-temperature su-pergene, not high-temperature hypogene. However, the ori-gin of the causative fluid, its salinity, age, and direction ofmovement are presently under investigation.

The metallurgical implications of the copper-bearing mi-croinclusions are that to expose them on fragment surfaceswould require fine grinding to a grain size well below 400


0361-0128/98/000/000-00 $6.00 417

FIG. 18. Plots of Cl vs. Cu of clays (A-C; the arrow indicates the trend ifincluded atacamite would be responsible for the Cu content of the smectites;the vertical population in A is shown expanded in B) and CuO vs. SiO2 (D;the arrow gives corresponding trend for chrysocolla). See Figure 11 for smec-tite abbreviations.

mesh. Secondly, besides the inordinate cost of such finegrinding, chemical extraction of microinclusion copper fromfeldspar host phases would entail extremely high acid con-sumption, since increased hydrolysis of feldspars by chemicalreaction with acid would result and cause neutralization.

In the case of Cu smectites, since the copper occurs instructurally bound, nonexchangeable, noninterlayer, octahe-dral sites, any process envisioned to remove the copper wouldnecessarily have to avoid ion exchange as a recovery mecha-nism. Alternatives to ion exchange could involve destabilizingthe smectite host, perhaps by its conversion to kaolinite whichmay only incorporate copper or other divalent cations in neg-ligible amounts.

The relevancy of process mineralogy guided by a geologicappreciation of ore genesis mechanisms has never beforebeen as important as it is today, because heap leaching tech-nology is begining to supercede pyrometallurgical smelting asthe extraction method of choice worldwide. Atmosphericemissions are acquiring unprecedented environmental con-sideration and what acid is recovered from stacks is recycledand used in leaching.

AcknowledgementsThis paper is dedicated in memory of our former professors

who taught a group of classmates petrology at the University

418 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 418

TABLE 5. Chemical Analyses of Cu-Bearing Smectitic Clay Minerals from the Oxidation Zone of Three Deposits

Rodomiro Tomic1 Potrerillos2 Komsomolsk region3Deposit Chile Chile Russia

Sample RT-13014 DDH-3758 All CC1 Yakhontovite

Analysis Vg, 9 Vg2, 29 Mean SD a1 (14) a2 (9) b1 (8) b2 (1) 1 2 3

SiO2 43.66 51.04 53.49 2.37 47.80 48.91 36.65 35.71 48.83 48.62 53.94Al2O3 9.55 14.75 21.76 4.84 3.41 0.49 16.00 12.69 0.30 0.42 0.08FeO 3.03 4.68 5.77 2.42 28.81 33.62 11.84 16.66 6.51 9.96 10.34MgO 0.61 0.80 1.25 0.59 1.11 0.73 21.07 18.63 10.15 1.39 6.11CaO 0.98 0.94 1.07 0.34 2.81 2.85 0.70 1.15 3.46 1.56 2.58Na2O 1.23 0.81 0.83 0.71 0.00 0.00 0.17 0.13 0.05 K2O 0.19 0.72 0.34 0.29 0.29 0.13 0.08 0.14 0.12 0.06 CuO 28.75 14.27 3.50 3.91 3.77 1.27 1.49 2.89 18.59 25.99 14.95Sum 88.00 88.00 88.00 88.00 88.00 88.00 88.00 88.00 88.00 88.00Cu 22.97 11.40 2.79 3.12 3.01 1.01 1.19 2.31 14.85 20.76 11.94

At. proportion based on 22 oxygensSi 7.29 7.68 7.50 0.31 7.88 8.15 5.56 5.64 7.86 8.24 8.49Al(IV) 0.71 0.32 0.50 0.31 0.12 2.44 2.36 0.06 0.00 0.00(IV) 8.00 8.00 8.00 8.00 8.15 8.00 8.00 7.92 8.24 8.49

Al(VI) 1.17 2.29 3.08 0.47 0.54 0.10 0.42 0.01 0.00 0.08 0.01Fe 0.42 0.59 0.68 0.30 3.97 4.69 1.50 2.20 0.88 1.41 1.36Mg 0.15 0.18 0.26 0.12 0.27 0.18 4.76 4.39 2.44 0.35 1.43Cu 3.63 1.62 0.38 0.46 0.47 0.16 0.17 0.35 2.26 3.33 1.78(VI) 5.37 4.69 4.41 0.21 5.26 5.13 6.86 6.94 5.57 5.18 4.59

Ca 0.18 0.15 0.16 0.05 0.50 0.51 0.11 0.19 0.60 0.28 0.43Na 0.40 0.23 0.22 0.19 0.00 0.00 0.05 0.04 0.02 0.00 0.00K 0.04 0.14 0.06 0.05 0.06 0.03 0.02 0.03 0.02 0.01 0.00

At. proportion based on 28 oxygensSi 9.28 9.77 9.55 0.40 10.03 10.38 7.08 7.18 10.01 10.49 10.80Al 2.39 3.33 4.56 0.91 0.84 0.12 3.64 3.01 0.07 0.11 0.02Fe 0.54 0.75 0.87 0.38 5.06 5.96 1.91 2.80 1.12 1.80 1.73Mg 0.19 0.23 0.33 0.15 0.35 0.23 6.06 5.59 3.10 0.45 1.82Cu 4.62 2.06 0.49 0.59 0.60 0.20 0.22 0.44 2.88 4.24 2.26Ca 0.22 0.19 0.20 0.07 0.63 0.65 0.14 0.25 0.76 0.36 0.55Na 0.51 0.30 0.28 0.24 0.00 0.00 0.06 0.05 0.02 0.00 0.00K 0.05 0.18 0.08 0.06 0.08 0.04 0.02 0.04 0.03 0.02 0.00

interlayer 0.78 0.67 0.57 0.44 0.71 0.68 0.23 0.33 0.81 0.38 0.55 noninterlayer 17.02 16.14 15.79 0.27 16.88 16.90 18.91 19.02 17.17 17.08 16.64

1 This study; Vg and Vg2 are green veinlets (see Table 3); all = average with standard deviation for the 142 smectite analyses2 A Cu porphyry deposit 460 km south of Radomiro Tomic; a1 = smectite, a2 = Fe-rich smectite, b1 = mixedlayer smectite-chlorite, and b2 = Fe-rich

mixed-layer smectite-chlorite (number of averaged analyses in parentheses); energy dispersive spectroscopy data; Levi (1997)3 A Cu-Sn deposit containing yakhontovite, a Cu smectite; electron microprobe data from Postnikova et al. (1986)

of California, Berkeley. Charles Meyer and Frank Turnerwere rare teachers whose skills with the microscope and ded-ication to education serve as an inspiration to us to solve prac-tical problems in industry. Through our professional careerswe have been guided by Chucks encouragement: If all elsefails, just look at it.

We are grateful for very helpful reviews by Eugene Ilton, A.Decarreau, and John Dilles. The composite samples were se-lected by Gonzalo Rojas and checked by Patricio Cuadra forsuitability. John Donovan assisted in the electron microprobework.

REFERENCESBailey, S.W., 1991, Chlorites: Structures and crystal chemistry: Reviews in

Mineralogy, v. 19, p. 347403.Bettison, L.A. and Schiffman, P., 1988, Compositional and structural varia-

tions of phyllosilicates from the Point Sal ophiolite, California: AmericanMineralogist, v. 73, p. 6276.

Brimhall, G., Cunningham, A.B., and Stoffregen, R., 1984, Zoning in pre-cious metal distribution within base metal sulfides: A new lithologic ap-proach using generalized inverse methods: ECONOMIC GEOLOGY, v. 79, p.209226.

Brimhall, G.H, and Arcuri, T., 1998, Controls on supergene and exotic min-eralization at the Radomiro Tomic mine, Chile: Santiago, Chile,CODELCO, Unpublished progress report, 20 p.

Cuadra C, P., and Rojas S., G., 2001, Oxide mineralization at the RadomiroTomic porphyry copper deposit, northern Chile: ECONOMIC GEOLOGY, v.96, p. 387400.

Decarreau, A., Grauby, O., and Petit, S., 1992. The actual distribution of oc-tahedral cations in 2:1 clay minerals: Results from clay synthesis: AppliedClay Science, v. 7, p. 147167.

Gven, N., 1991, Smectites: Reviews in Mineralogy, v. 19, p. 497559.International Centre for Diffraction Data, 1993, Mineral diffraction file

(Hanawalt, Fink and chemical search manuals and powder diffraction filessets): Swarthmore, Pennsylvania.

Jambor, J.L., and Puziewicz, J., 1991, New mineral names: American Miner-alogist, v. 76, p. 665671.

Levi, B., 1997, Estudio SEM-EDS y DRX de fases componentes de la mues-tra CC-1: Santiago, Chile, CODELCO, unpublished report, May 26, 16 p.


0361-0128/98/000/000-00 $6.00 419

FIG. 19. Comparison of Cu-bearing smectitic clay minerals in samples from the oxidation zones of Radomiro Tomic (RT,all analyses of this study) and two other deposits: the Potrerillos Cu porphyry (CC-1 Smt = smectite, CC-1 mix-l. = mixed-layer smectite-chlorite; Levi, 1997) and a Russian Cu-Sn deposit (Yakh. = yakhontovite, a Cu smectite; Postnikova et al.,1986). The legend of the diagrams is given in D (see Figs. 12-13 for the chlorite and smectite fields).

Mosser, C., Mestdagh, M., Decarreau, A., and Herbillon, A.J., 1990, Spec-troscopic (ESR, EXAFS) evidence of Cu for (Al-Mg) substitution in octa-hedral sheets of smectites: Clay Minerals, v. 25, p. 271282.

Mosser, C., Mosser, A., Romeo, M., Petit, S., and Decarreau, A., 1992, Nat-ural and synthetic copper phyllosilicates studied by XPS: Clays and ClayMinerals, v. 40, p. 593599.

Paquet, H., Colin, F., Duplay, J., Nahon, D., and Millot, G., 1987, Ni, Mn,Zn, Cr-smectites, early and effective traps for transition elements in super-gene ore deposits, in Rodriguez-Clemente, R., and Tardy, eds., Geochem-istry and mineral formation in the Earth surface: Consejo Superior de In-vestigaciones Cientficas [Spain]-Centre National de la RechercheScientifique [France], p. 221229.

Pons, J.-C., Parra, M., Ferragne, A. and Latouche, C., 1989, Caractristiquesdes argiles hydrothermales de la MartiniquePet Antilles Franaises: Ap-plied Clay Science, v. 4, p. 307325.

Postnikova, V.P., Tsipurskij, S.I., Sidorenko, G.A., and Mokhov, A.V., 1986,Yakhontovitea new copper-bearing smectite: Mineralog Zhurnal, v. 8, p.8084 (in Russian).

Rybicka, E.H., and Jedrzejczyk, B., 1995, Preliminary studies on mobilizationof copper and lead from contaminated soils and readsorption on competingsorbents: Applied Clay Science, v. 10, p. 259268.

Schiffman, P., and Fridleifsson, G.O., 1991, The smectite-chlorite transitionin drill hole NJ-15 Nesjavellir geothermal field, Iceland: XRD, BSE andelectron microprobe investigations: Journal of Metamorphic Geology, v. 9,p. 679696.

Schmidt, S.T., and Robinson, D., 1997, Metamorphic grade and porosity andpermeability controls on mafic phyllosilicate distributions in a regional ze-olite to greenschist facies transition of the North Shore Volcanic Group,Minnesota: Geological Society of America Bulletin, v. 109, p. 683697.

Thorez, J., 1976, Practical identification of clay minerals: Editions Lelotte,Belgium, Dison, 90 p.

420 BRIMHALL ET AL.

0361-0128/98/000/000-00 $6.00 420

Documents

Brimhall Et Al-2001