Zwahlen 2014

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/262563233
The Porphyry Cu-(Mo-Au) Deposit at Altar (Argentina): Tracing Gold

Distribution by Vein Mapping and LA-ICP-MS Mineral Analysis
Article in Economic Geology · May 2014

DOI: 10.2113/econgeo.109.5.1341
CITATIONS READS
18 810
5 authors, including:
Carmen Zwahlen Thomas Wagner

Luleå University of Technology RWTH Aachen University
6 PUBLICATIONS 36 CITATIONS 111 PUBLICATIONS 2,761 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Element Mobility in Hydrothermal Systems View project
GEMS TM: Gibbs Energy Minimization Software for Thermodynamic Modelling View project
All content following this page was uploaded by Thomas Wagner on 23 July 2014.
The user has requested enhancement of the downloaded file.

©2014 Society of Economic Geologists, Inc.
Economic Geology, v. 109, pp. 1341–1358
The Porphyry Cu-(Mo-Au) Deposit at Altar (Argentina):

Tracing Gold Distribution by Vein Mapping and LA-ICP-MS Mineral Analysis
Carmen Zwahlen,1,† Stefania Cioldi,1 Thomas Wagner,1,* Roger Rey,2 and Christoph Heinrich1
1 ETH Zurich, Institute of Geochemistry and Petrology, Clausiusstr. 25, CH-8092 Zurich, Switzerland
2 Peregrine Metals LTD., Avda. El Bosque Norte 0123, Oficina 604, Las Condes, Santiago, Chile
Abstract
The Altar porphyry Cu-(Mo-Au) deposit occurs in the Miocene to early Pliocene copper belt of the high
Andes of western Argentina, close to the Chilean border. A cluster of late Miocene porphyries intruded a more
extensive complex of early Miocene rhyolitic to andesitic volcanic and subvolcanic rocks. Potassic alteration
in the Central Porphyry was partially overprinted by feldspar-destructive alteration and finally by argillic to
advanced argillic alteration. Early stockwork veins have been cut successively by quartz-pyrite veins, tourmaline
veins, and enargite veins. Based on correlation analysis of assay results and mapped abundances of these four
vein types, we estimate that approximately 11 to 26% of the copper in the Altar orebody is associated with enar-
gite veins, whereas the remainder is associated with the early stockwork veining and potassic alteration. Gold
concentration is low compared with typical Cu-Au porphyries of the Andean back-arc region, but higher than
in the giant Miocene deposits of Chile, at an average Au/Cu ratio of 0.14 × 10–4 by weight across the Central
Porphyry orebody at Altar. Gold is dominantly associated with chalcopyrite in the domain of stockwork veining
and potassic alteration. Small-scale assays in combination with petrographic observations and laser ablation-
inductively coupled plasma-mass spectrometry microanalyses show that a significant fraction of the gold in
the stockwork veins is enclosed in pyrite as submicroscopic particles attached to inclusions of a Bi-Te-Pb-Ag–
enriched Cu-Fe sulfide phase. Microscopic grains of native gold also occur along grain boundaries between
pyrite and chalcopyrite, and rarely as larger particles in enargite veins.
Introduction by substitution into bornite and, to a minor extent, into chal-

Highly variable gold contents occurring in porphyry copper copyrite (Simon et al., 2000; Kesler et al., 2002). However, the
orebodies are well documented by bulk assay data, but our temperature range of Cu-Fe sulfide precipitation in porphyry
knowledge and understanding of gold distribution among dif- ores is commonly too low to accommodate the entire gold
ferent mineral phases within porphyry ore zones are incom- content of a deposit in Cu-Fe sulfide solid solution (Land-
plete (Kesler et al., 2002). Some gold occurs as discrete twing et al., 2010). Alternatively, systematic differences in Cu/
grains, commonly associated spatially with Cu-Fe sulfides on Au ratios have been related to the depth of formation of dif-
all scales (Kesler et al., 2002). Gold can also occur as invisible ferent porphyry deposits, where deeper deposits are charac-
gold in sulfide minerals (e.g., Simon et al., 2000). A practical terized by a higher Cu and Mo content compared to shallower
definition for invisible gold is that it is present within a host ones with a higher Au/Cu ratio (Murakami et al., 2010). Metal
phase at a scale that is not detectable by optical and scan- separation may occur due to fluid phase separation (the tran-
ning electron microscopy; it can either be chemically bound sition from intermediate-density fluid to liquid-vapor coex-
as a true solid solution in the crystal structure or may occur istence) at different depth, resulting in Cu-Au fractionation
as nanoparticles with a size smaller than ~100 nm (Cook and by selective metal precipitation, depending on the extent of
Chryssoulis, 1990). decompression (Landtwing et al., 2010).
Three major factors contribute to the degree of gold enrich- 3. The third proposed mechanism is a later high- or low-
ment in porphyry deposits: temperature hydrothermal overprint on the porphyry-style
mineralization which may selectively remove or add cop-
1. The initial composition of the magma influences its abil- per or gold, thereby contributing to variations in the Cu/Au
ity to concentrate and transport metals in silicate and/or sul- ratio in porphyry copper systems (Kesler et al., 2002). The
fide melts (Ballard et al., 2002; Richards, 2003), which in turn magnitude of invisible gold incorporation into sulfides and its
affects the partitioning of copper and gold into magmatic- postdepositional liberation has been described for arsenian
hydrothermal fluids (Ulrich et al., 1999; Halter et al., 2002; pyrite and arsenopyrite in metamorphic and sediment-hosted
Frank et al., 2011). epithermal ore deposits (Richards and Kerrich, 1993; Reich
2. Precipitation mechanisms influence the proportion of et al., 2005; Wagner et al., 2007; Large et al., 2009), indicat-
Au and Cu through a range of mechanisms. At high tempera- ing an essential requirement of significant arsenic substitution
ture, variable amounts of gold can be incorporated as a solid in pyrite to allow incorporation of invisible gold (Fleet and
solution component into Cu-Fe sulfide minerals, particularly Mumin, 1997; Deditius et al., 2008; Cook et al., 2009). How-
ever, only few studies have documented mineralogy, trace ele-
† Corresponding
ment chemistry of sulfides, and deposit-scale distribution of
author: e-mail, zwahleca@gmail.com
*Present address: Department of Geosciences and Geography, Division
Cu and Au in higher-temperature porphyry and epithermal
of Geology, University of Helsinki, P.O. Box 64 (Gustaf Hällströmin katu 2a), deposits (e.g., Arif and Barker, 2004; Cook et al., 2011; Deyell
FI-00014 Helsinki, Finland. and Hedenquist, 2011; Gregory et al., 2013).
Submitted: August 19, 2012
0361-0128/14/4232/1341-18 1341 Accepted: August 14, 2013
1342 ZWAHLEN ET AL.
The world’s greatest concentration of porphyry copper the host rocks associated with the Central Porphyry and the
deposits occurs in several orogen-parallel belts in the South related alteration styles, vein types, and mineral and metal
American Andes. These metallogenic belts formed at discrete distribution. A second stock, Altar East, is currently being
times from the Cretaceous to the Pliocene (Sillitoe and Per- explored but will not be described in this paper. Because gold
elló, 2005). Central Chile (29°–32°S) hosts several giant, late is present in small but economically significant concentrations,
Miocene to Pliocene deposits of the Cu-Mo type with low Au its spatial distribution at all scales is particularly important for
contents (Sillitoe and Perelló, 2005). Prominent examples the development of a future mine. This paper describes the
include Los Pelambres-El Pachon (890 Mt at 0.61% Cu, main vein types in geologic context, and then analyzes the gold
0.01% Mo, and 0.015 ppm Au; Mutschler et al., 2010; Fig. distribution at the deposit scale using a statistical treatment of
1), Rio Blanco-Los Bronces (5 Bt at 1% Cu and 0.02% Mo; assay data and vein logging from an extensive recent explo-
Mutschler et al., 2010), and El Teniente (12 Bt at 0.92% Cu, ration program. By combining these deposit-scale data with
0.02% Mo, and average Au grade of 0.002 ppm; Mutschler detailed vein petrography and microanalysis of trace elements
et al., 2010). Porphyry deposits of similar age in Argentina, in vein minerals by laser ablation-inductively coupled plasma-
including El Pachon, Altar, Piuquenes, Los Azules, and mass spectrometry (LA-ICP-MS), we can identify different
Bajo de la Alumbrera (780 Mt at 0.52% Cu, 0.04% Mo, and stages of gold mineralization and can approximately quantify
0.67 ppm Au; Mutschler et al., 2010), are located at variably the location of gold in relation to Cu, As and the principal evo-
greater distance from the location of the present-day subduc- lution stages of the currently best explored orebody at Altar.
tion zone and are variably enriched in gold (Sillitoe and Per-
elló, 2005). A recent study of the Altar area has integrated this Regional Geologic Setting
deposit into a regional geodynamical model and assigned it to The Central Andes evolved above continental basement
the group of generally gold poor Miocene deposits in Chile shaped by the Gondwanan orogeny in the late Paleozoic
(Maydagan et al., 2011), even though Altar has a noticeably (Ramos, 1988; Heredia et al., 2002), starting with basins that
higher Au/Cu ratio (802 Mt at 0.42% Cu, 0.06 ppm Au, and were filled with silicic to intermediate volcanic rocks and clas-
1.4 ppm Ag at a cutoff rate of 0.3% Cu equiv; Simpson et al., tic sediments during Mesozoic extension (Ramos, 1988; Fig.
2011) compared to the nearby Chilean deposits such as Los 1). When spreading of the South Atlantic Ocean was initiated
Bronces and Los Pelambres (Mutschler et al., 2010). in the Late Cretaceous, convergence rates and directions
In this study, we present a geologic description of the por- changed along the Chilean continental margin and initiated
phyry copper-gold deposit at Altar (Argentina), focusing on the main compressional stage of the Andean orogeny. Some of
70°45’W 70°W
31°S
Q Conglomerate,
sandstone
late
Dacitic and andesitic
porphyries
Clastic sedimentary
ARGENTINA rocks, tuff, evaporite
CHILE
Cenozoic
Mi
Calingasta Volcanics,
early
sedimentary rocks
Pachon Formation,
bimodal andesite and
rhyolite, clastic
sedimentary rocks
Altar Pg
Andesitic to dacitic
porphyries
Gabbro, granodiorite
Cr Andesitic volcanics
Mesozoic
Sandstone,
conglomerate
Granodiorite,
Los Pelambres El Pachon monzogranite
Clastic sedimentary
J rocks, limestone,
evaporite
Tr Silicic to intermediate
volcanics
Paleoz.
P
Granodiorite
32° S
10 km C Sandstone,
conglomerate
Fig. 1. Geologic map of the Central Andes obtained by integrating map sheets from the Secretaria de Industria, Comercio
y Minería et al. (1999) and Instituto de Investigaciones Geológicas (1978) with recent mapping data by Perez et al. (2010).
Abbreviations: C = Carboniferous, Cr = Cretaceous, J = Jurassic, Mi = Miocene, P = Permian, Paleoz. = Paleozoic, Pg =
Paleogene, Q = Quaternary, Tr = Triassic.
GOLD DISTRIBUTION AT THE PORPHYRY Cu-(Mo-Au) DEPOSIT, ALTAR, ARGENTINA 1343
the preexisting extensional faults were reactivated as reverse indicated resources are 802 million tonnes averaging at
faults and thrusts (Heredia et al., 2002). The evolution of the 0.42% Cu and 0.059 ppm Au at a cutoff grade of 0.3% Cu
early Andes ended with slow plate convergence and low mag- equiv in and around the Central Porphyry (Simpson et al.,
matic activity in the Oligocene (Kay et al., 2005). 2011). A second porphyry body of similar outcrop size is cur-
The breakup of the Farallon plate into the Nazca and Cocos rently under exploration (Altar East; Fig. 3). These grades are
plates since the early Miocene led to a change in convergence based on 58,642 m of drill core from 140 drill holes. The host-
direction and an increase in the subduction rate, promoting rock lithology, successive vein types, and mineralization char-
enhanced uplift (Kay and Mpodozis, 2002) with extensive acteristics were first documented by Almandoz et al. (2005),
bimodal volcanism in continental basins (Fig. 1, Pachon For- based on exploration data and samples from Rio Tinto.
mation). Shallowing of the subduction zone commenced in
the Chilean flat slab segment (28°–33°S) about 18 m.y. ago Host rock lithology
and was accompanied by crustal thickening. This resulted in Early Miocene volcanic and subvolcanic rocks exposed in
a change in the style of volcanism from andesitic stratovolca- the Altar cirque have been intruded by late Miocene porphy-
noes to dacitic dome complexes and upper-crustal intrusions ries (Figs. 3–5). The early Miocene volcanic complex consists
in the late Miocene (Kay and Mpodozis, 2002). The gap in of andesitic to dacitic lava flows and domes surrounded by
extrusive volcanic activity established in the Altar region is rhyolitic tuff that has been dated at 20.8 ± 0.3 Ma (Maydagan
associated with a change from normal subduction to a com- et al., 2011). Andesitic rocks occur as two distinct varieties,
pressive state in the later Miocene (Kay and Mpodozis, 2002). the age relationships of which are not yet clear. The oldest
The late Miocene magmatic activity correlates with the subvolcanic rocks cropping out regionally are of andesitic com-
duction of the Juan Fernandez ridge beneath the Altar area position (Fig. 5A). These dark brown to black fine-grained
(Yanez et al., 2001; Richards, 2003; Cooke and Hollings, 2005; porphyritic andesitic to dacitic lava flows typically contain
Hollings et al., 2005; Rosenbaum et al., 2005; Maydagan et plagioclase phenocrysts with diameters of about 1 mm. The
al., 2011). In recent studies of the igneous geochemistry and second variety is a porphyritic andesite that has been inter-
geochronology of the Altar area, Maydagan et al. (2011, 2014) sected by drilling in the center of the deposit. This lithology
argue that trench-ridge collision and the subduction of the has abundant plagioclase phenocrysts up to 1 cm in size set
Juan Fernandez ridge in the study area produced changes in in a fine-grained black matrix, indicating rapid quenching
the stress regime, eventually leading to the rising of stored in a volcanic or shallow subvolcanic environment. Textures
magma into the upper crust (Richards, 2003). Further flatten- indicating magma mingling with the overlying rhyolite sug-
ing of the subduction zone brought the volcanism in the Main gest that the andesitic dome or subvolcanic intrusion is coeval
Cordillera to an end at about 9 Ma (Kay and Mpodozis, 2002). with the early Miocene rhyolite (Fig. 5C). However, drilling
at Altar East has intersected rhyolite clasts within an andesite
Altar Deposit Geology unit, indicating that some andesite emplacement postdates
the rhyolite and may be part of the late Miocene mineralizing
Locality and exploration history Central Porphyry.
The Altar Cu-(Mo-Au) porphyry deposit (S31°28'15", Rhyolite surrounds the Central Porphyry and the andesite
W70°29'5") is located about 8 km to the east of the Chilean porphyry in the center of the Altar cirque and has truncated
border and approximately 23 km to the north of Xstrata’s the andesitic rocks in the east and the west (Fig. 5B). Rhyo-
El Pachon project (Fig. 1; Sillitoe and Perelló, 2005). Altar lite also crops out on ridges and crests around Altar, whereas
is exposed at altitudes of 3,300 to 4,100 m above sea level andesite is restricted to the valleys, suggesting a volcanostrati-
(a.s.l.). It is located in a postglacial cirque at the head of a graphic contact between both rock types. Andesite clasts are
valley draining into the Pantanosa River and is surrounded by common in the rhyolite, demonstrating that the rhyolite is
talus-covered mountains (Fig. 2). coeval or younger than at least part of the andesitic complex.
Recent exploration at Altar was commenced by Rio Tinto The rhyolite has a well-developed porphyrytic texture with
in 2001. Peregrine Metals Ltd. then explored the site from approximately 30 vol % phenocrysts (Fig. 5B). Feldspar and
2005 until its takeover by Stillwater in 2011. Measured and subrounded quartz phenocrysts occur within a homogeneous
microcrystalline groundmass. The matrix color varies from
light pink to dark gray depending on the amount of finely dis-
W seminated hematite. Fiamme textures have been observed in
some drill cores and are interpreted to indicate that the rhyo-
litic sequence contains ignimbrites.
There are a clear hiatus and distinct textural difference
between the bimodal volcanic rocks and the late Miocene
N
porphyry intrusions, which are volumetrically small and have
S
no preserved extrusive equivalents in the region. This hiatus
Central porphyry is regionally correlated with a change from extensional basin
deposition to basin inversion, with increased compression
leading to storage of magma in the mid-crust and emplace-
ment of intrusive, phenocryst-rich porphyries in the upper
Fig. 2. Photograph of the Altar cirque seen from Altar East, showing the crust (Richards, 2003; Maydagan et al., 2011). The porphyry
outcrop location of the Central Porphyry. suite contains several mineralized and nonmineralized stocks
1344 ZWAHLEN ET AL.
6518000m N
2358500m E
2361500m E
70°29“1‘W N
111 110 108 94
90 88 91 103 95 114
131 128 86 98 124
93 121 123 84 127 122 130 Altar East

117
60 134 140
Central porphyry
82
120 47 85 80 77 135 116
89
115 50 76 92 102 107
110 62 46 49 43 57 63 68 61
31°28“55‘S 14
133 101 66 53 51 75 87 97 Late Miocene

96 71 56 69 79 129 Central porphyry Potassic alteration
6516500m N
136 67 132 81 125
Altar East porphyry Propylitic alteration
Early Miocene Sericitic alteration
139 138 126 137
Rhyolite Advanced argillic alt.
300 meters Andesite Fault Silica ledge
Fig. 3. Deposit-scale map illustrating distribution of the most important rock types (colors), major alteration types (pat-
terns), faults, and exploration drill holes (points, numbers). Rocks shown as darker colors have been mapped in outcrop,
whereas lighter colors indicate inferred geology based on float mapping. The white numbers are drill holes from which small-
scale assay samples were taken.
of plagioclase-rich andesite to dacite porphyry, which were the Central Porphyry have been constrained from detailed
placed into a tentative intrusion sequence by Maydagan et field mapping, drill core logging, and assay data (Figs. 3–4).
al. (2014). The mineralized Central Porphyry is exposed in Hypogene copper mineralization at Altar consists mainly of
the central part of the deposit (Fig. 5D). The texture is dis- chalcopyrite with smaller amounts of bornite, pyrite, molyb-
tinctly porphyritic with 60 vol % plagioclase phenocrysts up denite, and magnetite. The distribution of high-grade copper
to 1 cm in diameter, 8 vol % of former amphiboles converted is closely tied to the Central Porphyry (Fig. 4), with some cop-
to secondary biotite and chlorite, and 2 vol % of primary bio- per occurring in the surrounding andesite and rhyolite. Gold
tite. The holocrystalline matrix consists mainly of quartz and is also mainly enriched within the Central Porphyry and spa-
plagioclase. The Central Porphyry was subdivided into petro- tially correlates with the intermediate- to high-grade copper
graphically similar northern and southern intrusions by May- zone, particularly in the upper part of the deposit. Elevated
dagan et al. (2014), because the southern intrusion includes a molybdenum concentrations occur farther out than high-
few distinctly younger zircon analyses in an otherwise overlap- grade copper and gold, forming an outer shell around the cen-
ping age population. On the basis of interpreting the youngest tral copper-gold zone. Most of the molybdenite is hosted in
zircons as the best estimate for intrusion age (von Quadt et al., the rhyolite. Potassic alteration affected the Central Porphyry,
2011), Maydagan et al. (2011, 2014) reported distinct U/Pb andesitic intrusive rocks, and parts of the rhyolite (Figs. 3, 4C,
ages of 11.13 ± 0.26 Ma for the northern intrusion and 10.35 ± D). Maps and cross sections demonstrate that potassic altera-
0.32 Ma for the southern one. We question the significance of tion and high copper grades largely overlap (Fig. 4D).
this age difference because the published LA-ICP-MS analy- A domain of propylitic alteration surrounds the potassic
ses did not allow removal of grain portions that may have suf- alteration zone (Fig. 3). The propylitically altered rocks con-
fered Pb loss. tain epidote and chlorite that mostly formed via breakdown
The Central Porphyry has been crosscut by pebble dikes of igneous biotite and amphibole, but also partially replaced
up to 3 m in width that contain rhyolite clasts and quartz vein some feldspars.
fragments up to 10 cm in diameter. The pebble dikes have Late-stage feldspar-destructive alteration (sericitic to argil-
fine sandy matrix and sharp subvertical contacts, suggesting a lic) overprinted the potassic alteration to variable degrees
late timing with respect to the mineralizing intrusion. and affected the entire porphyry and wall rocks beyond the
maximum lateral extent of drilling (Fig. 4C, D). In this lower-
Alteration and mineralization temperature alteration zone that formed under more acid
The characteristics and spatial distributions of altera- conditions, tourmaline, chalcocite, chalcopyrite, and pyrite
tion assemblages and mineralization types in and around make up the main ore mineral assemblage, occurring in
N S W E
Intersection
Intersection
2.359.000m E
2.360.400m E
6.517.200m N
6.516.200m N
31°28“55‘S 70°29“1‘W
3800m 3800m
3600m 3600m
3400m 3400m
3200m 3200m
3000m 3000m
2800m 2800m Central Porphyry

Central Porphyry Andesite
Andesite Rhyolite
Rhyolite Leached Central Porphyry
Leached Central Porphyry Leached Rhyolite
2600m Leached Rhyolite A 2600m Leached Andesite B
N S W E
6.517.200m N
6.516.200m N
2.359.000m E
2.360.400m E
31°28“55‘S 70°29“1‘W
3800m 3800m
3600m 3600m
3400m 3400m
3200m 3200m
3000m 3000m
Potassic alteration Potassic alteration

Feldspar-destructive alteration Feldspar-destructive alteration
2800m Cu > 0.6 wt.% 2800m Cu > 0.6 wt.%
Cu > 0.3 wt.% Cu > 0.3 wt.%
Cu > 0.2 wt.% Cu > 0.2 wt.%
Au > 0.4 ppm Au > 0.4 ppm
Au > 0.2 ppm Au > 0.2 ppm
2600m Au > 0.1 ppm C 2600m Au > 0.1 ppm D
Fig. 4. Representative cross sections through the Central Porphyry at Altar. (A) N-S geologic section. (B) E-W geologic
section. (C) N-S alteration and mineralization section. (D) E-W alteration and mineralization section.
veins and as disseminations. Advanced argillic alteration was Supergene mineralization at Altar is concentrated in a sub-
observed in Altar East and is characterized by an assemblage horizontal zone approximately 150 m thick. The enrichment
of native sulfur, kaolinite, or dickite, barite, and alunite, fill- zone is located at elevations of 3,350 to 3,500 m above sea
ing open space in vuggy quartz formed by complete leaching level, whereas the present-day surface is between 3,500 and
of feldspar (Fig. 3). The hypogene copper grade largely over- 4,100 m, indicating that supergene enrichment occurred prior
laps the distribution of the potassic alteration and decreases to recent glacial activity (Fig. 4C, D). Subhorizontal gypsum
upward into the zone of feldspar-destructive alteration. This veinlets occur in most drill cores and are probably related to
results in a pronounced low-grade zone between the deep the supergene overprint; however, the origin of this vein type
ore zone and the near-surface zone of supergene enrichment is not clear. Supergene jarosite, goethite, and hematite veins
(Fig. 4D). and fractures have overprinted sulfide veins.
1346 ZWAHLEN ET AL.
A B following Almandoz et al., 2005), which consists of four stages

and 11 substages, based on similarities in relative timing and/
or mineral assemblages. The four paragenetic stages relate to
specific alteration assemblages and include (1) early quartz
stockwork veins related to potassic alteration, (2) quartz-pyrite
veins and (3) tourmaline veins, both associated with feldspar-
destructive alteration, and (4) late enargite veins associated
with argillic alteration (Fig. 6). Relative vein densities were
determined by dividing drill cores into intervals of 2-m length,
with the relative abundance of each vein type extracted from
1cm 1cm the logging database of Peregrine Metals Ltd. Relative vein
densities ranging from one (rare) to five (very abundant,
C D approximating 30 vol % or more) have been attributed to the
four dominant vein types within an interval.
Stage 1 stockwork veins contain quartz, chalcopyrite, and
minor amounts of pyrite, bornite, anhydrite, and molybdenite
(Fig. 7A). Some quartz veins contain irregularly distributed
sulfides (stage 1A; Fig. 7A), whereas others have a central
seam of sulfides (stage 1C; Fig. 7B). Stage 1A and 1C veins
are similar to A and B veins described by Gustafson and Hunt
(1975) at the El Salvador porphyry Cu-Mo deposit, Chile.
1cm Stage 1B molybdenum-quartz veins at Altar crosscut stage 1A
1cm
veins, but in turn they are crosscut or reopened by stage 1C
Fig. 5. Hand specimen photographs of main host rock types. (A) Andesite veins. Two subgroups of stage 1B veins appear to have formed
(drill core ALD-46, 400 m). (B) Rhyolite (surface sample, east of Central Por- at the same time, because no crosscutting relationships could
phyry). (C) Andesite mingling with rhyolite (drill core ALD-55, 299 m). (D) be observed (Fig. 7C, D). Stage 1D gray quartz veins have a
Central Porphyry with potassic alteration overprint, indicated by brown spots
that replaced former amphibole (drill core ALD-43, 880 m).
distinct K-feldspar halo and are filled with whitish gray quartz,
pyrite, and chalcopyrite (Fig. 7E, F). Stage 1D veins only
occur at depth in the center of the porphyry stock.
Vein Types and Metal Correlation Stage 2 quartz-pyrite veins are associated with feldspar-
destructive alteration. They consist essentially of quartz and
Macroscopic features of main vein types pyrite with minor amounts of chalcopyrite, are surrounded
A total of 39 vein types were identified at Altar by Pere- by different types of alteration halos, and show a consistent
grine Metals Ltd., based on vein mineralogy and textures. We sequence of emplacement (Fig. 7G, H). Stage 2A veins con-
have defined a simplified paragenetic scheme (Fig. 6, largely tain pyrite and quartz with minor amounts of chalcopyrite and
Vein stages Substages - vein mineralogy Potassic Feldspar-destructive Advanced argillic

alteration alteration alteration
1A Gray quartz with sulfides (py, cp, ±bn, ±anh)
1B Molybdenite - quartz / molybdenite - quartz with py and cp

Stage 1
Stockwork veins 1C Gray quartz with pyrite and chalcopyrite, with center line
1D White quartz with pyrite, chalcopyrite and feldspar halo
2A Quartz - pyrite (chalcopyrite) with silicic halo
2B Quartz - pyrite (chalcopyrite) with sericite - tourmaline halo

Stage 2
Quartz-pyrite veins 2C Pyrite
2D Milky quartz with pyrite
2E Hematite - magnetite with sulfides (± py, ±cp)
Stage 3
3 Tourmaline with sulfides (py, cp, cc)
Tourmaline veins
Stage 4
4 Enargite with pyrite, quartz and gypsum / anhydrite
Enargite veins
Time
Fig. 6. Paragenetic sequence diagram illustrating time relationships of the major vein types and the related wall-rock
alteration. Abbreviations: anh = anhydrite, bn = bornite, cc = chalcocite, cp = chalcopyrite, py = pyrite.
A B C D
1cm 2mm 5mm 5mm

E F G H
5mm 1cm 1cm 5mm
I J K
1cm 1cm 1cm

Fig. 7. Hand specimen photographs of major vein types. (A) Stage 1A stockwork veins (surface sample of the Central Por-
phyry). (B) Stage 1C stockwork veins with sulfides disseminated in the center, appearing after molybdenite veins (drill core
ALD-124, 78 m). (C) Stage 1B molybdenite-quartz vein (drill core ALD-68, 437 m). (D) Stage 1B molybdenite-sulfide vein
with molybdenite at the vein margin and sulfides in the center (drill core ALD-68, 443 m). (E) Stage 1D pyrite-quartz veins
with feldspar halo (drill core ALD-46, 336 m). (F) Stage 1D pyrite-quartz veins with feldspar halo (drill core ALD-132, 516
m). (G) Stage 2A pyrite-quartz vein with silicic halo cut by stage 2B quartz-pyrite vein with sericite halo (drill core ALD-49,
426 m). (H) Stage 2B quartz-pyrite veins with sericite-tourmaline halo (drill core ALD-49, 458 m). (I) Stage 3 tourmaline
vein cut by stage 2C pyrite vein (drill core ALD-64, 211 m). (J) Stage 2D milky quartz-pyrite veins cutting quartz-pyrite vein
with silicic halo (drill core ALD-46, 384 m). (K) Stage 4 late enargite-pyrite-quartz-gypsum vein (drill core ALD-64, 77.4 m).
have a silicic halo with minor sericite, locally with dissemi- Stage 4 enargite veins contain varying proportions of pyrite,
nated pyrite (Fig. 7G). These veins are crosscut by stage 2B quartz, and gypsum (probably after anhydrite; Fig. 7K). Stage
quartz-pyrite veins that also contain minor chalcopyrite, but 4 veins are most common at shallow depths and are gener-
have an inner sericite and an outer tourmaline halo (Fig. 7H). ally low in abundance but can reach up to 5 cm in thickness.
Stage 2C massive pyrite veins cross all previous vein types These veins have formed over a larger area than most other
(Fig. 7I). Some stage 2D milky quartz-pyrite veins have cut vein types and are associated with advanced argillic alteration.
stage 2A veins (e.g., Fig. 7J). The scarcity of stage 2D veins
precludes constraining their timing relationships in greater Correlation between assay data and mapped vein densities
detail. Stage 2E magnetite-hematite veins contain minor Data from 68 drill holes from the Central Porphyry logged
amounts of sulfides, crosscut stage 2A and 2B veins, and occur in the years 2009 to 2010 have been used to analyze corre-
in distal locations within the Altar cirque. lations between metal assays and vein types, amounting to a
Stage 3 tourmaline veins locally contain subordinate total database of 22´815 m of core assayed by Acme Analytical
amounts of sulfides and quartz (Fig. 7I), crosscut stage 2E Laboratories (Chile) by ICP-optical emission spectrometry
veins, and are more common in distal locations. They are also (OES) at 2-m intervals for Cu, Mo, Au, Ag, As, Al, B, Ba, Bi,
associated with feldspar-destructive alteration. Ca, Cd, Co, Cr, Fe, Hg, La, K, Mg, Mn, Na, Ni, P, Pb, S, Sb,
1348 ZWAHLEN ET AL.
Sr, Th, Zn, U, V, W, and Ti. The following analysis has been 3000 1.2
restricted to hypogene mineralization by excluding all inter- A B
vals logged as oxidized or dominated by supergene copper 2500 1.0
minerals (black chalcocite, Cu oxides, and carbonates), reduc-
ing the dataset to 18,528 m of drill core and 9,264 samples. 2000 0.8
The assay data have been plotted to test for correlations
As (ppm)
Au (ppm)
among the four ore-related elements Cu, Mo, Au, and As (Fig. 1500 0.6
8). A linear data array is apparent in Figure 8A, which shows
a clear positive correlation between Cu and As in many high-
1000 0.4
grade samples, conforming to the stoichiometric Cu/As ratio
of enargite, Cu3AsS4. Only few data points, which come from
samples of the transitional zone between the supergene and 500 0.2
hypogene mineralization, plot above the enargite line. Here,
arsenic probably occurs as so far unidentified arsenate miner- 0 0
0 0.4 0.8 1.2 1.6 2 0 500
als, whereas copper is lost in the weathering process (Musu
et al., 2007a, b; Lattanzi et al., 2008). Molybdenum shows no Cu (wt.%)
correlation with copper,
3000 consistent with the observation that 1.2
there is very little Mo inAhigh-grade copper ores whereas Mo B Assay data of Cu and As
Enargite: vein density 1
enrichment occurs2500 in an outer zone containing only low to 1.0 Enargite: vein density 2
moderate copper grades (Fig. 8B). There is a weakly positive Enargite: vein density 3
correlation (r2 = 0.05) between Au and Cu with an Au/Cu ratio Enargite: vein density 4
2000 0.8 Enargite: vein density 5
of about 0.1 ppm/wt % (Figs. 8C, 9B), consistent with the
As (ppm)
Au (ppm)
bulk resource average (0.14) within uncertainty. High-grade
1500 0.6
copper generally appears together with elevated gold grades,
but some data points extend to much higher gold grades. A
weak positive correlation
1000 (r2 = 0.03) is also visible between 0.4
Au and As (Au/As: 0.4 ppm/wt %), but the samples with the
higher gold grades at 500low copper are located above this trend 0.2
(Fig. 8D). Therefore, the highest gold grades are not associ-
ated with either high 0copper or high arsenic. 0
Assuming that the logging
0 of 0.4
vein densities
0.8 was 1.2
equally rep-
1.6 2 0 500 1000 1500 2000 2500 3000 3500
resentative for all vein types within the rock volume sampled
Cu (wt.%) As (ppm)
by the selected drill holes, the combination of vein density Fig. 9. Correlations between metal grades and vein densities. (A) Cu vs.
data with assay data allows a first order quantification of the As with superimposed enargite vein densities 1 to 5 indicated by circle size
(the red line with green band indicates enargite stoichiometry). (B) As vs. Au,
3000 300 with superimposed enargite vein densities 1 to 5.
A B
Mo (ppm)
As (ppm)
dominant ore minerals. In this analysis, assay data have been

correlated with the logged density of stage 1, 2, 3, and 4 veins.
1000 100
Vein densities were classified in a semiquantitative way, with
categories ranging from weak (1) to intensely veined (5). Fig-
0 0 ure 9A shows that all assay intervals with a high density of
0 1 2 3 0 1 2 3
Cu (wt.%) Cu (wt.%) enargite veins plot close to the enargite line. Due to the clus-
1.2 1.2
tering of most large symbols along this trend, representing
C D high vein density, we estimate that the minimum amount of
copper hosted by enargite is approximately 11% of the total
Au (ppm)
Au (ppm)
copper at Altar. Many samples (6,967 out of 9,264) contain

copper in excess of As compared to the enargite stoichi-
0.4 0.6 ometry, i.e., they plot below the enargite line in Figure 9A.
Assuming that arsenic occurs exclusively as enargite in these
0 0
samples, the excess copper allows calculating a maximum
0 1 2 3 0 1000 2000 3000 fraction of copper bound to enargite, which is about 26% of
Cu (wt.%) As (ppm) the total copper in the orebody. In other words, we estimate
Fig. 8. Summary of assay data (all data plotted in black, overlain by all that between 74 and 89% of the Cu contained in the orebody
samples below 3,400-m elevation, highlighted in green). (A) Cu vs. As. The is bound in Cu-Fe sulfides but that a significant remaining
red line indicates enargite stoichiometry. (B) Cu vs. Mo. (C) Cu vs. Au. The fraction is present in the form of enargite.
red line indicates an Au/Cu ratio of 0.1 ppm/wt %, obtained by fitting the
green data (i.e., exclusive of shallow samples subject to supergene overprint).
The amount of gold associated with enargite veins can be
(D) As vs. Au. The red line indicates a similarly fitted Au/As ratio of 0.4 ppm/ estimated by inspecting the intervals where at least one enar-
wt %. gite vein was mapped. A fraction of 16% of the total gold is
related to stage 4 enargite veins if all the gold in these intervals Cu budget (compare Fig. 11A, D, J), but abundant As/Cu
is exclusively bound to enargite. Alternatively, the gold frac- ratios in excess of enargite stoichiometry indicate partial mod-
tion can be calculated for the intervals falling into the band in ification by supergene Cu leaching and presence of residual
Figure 9A, where all copper is thought to be within enargite, arsenates in the oxidized zone. Between 3,400 and 3,250 m
which is 24% of the total gold. The absolute amount of gold a.s.l., some enargite veins are present, but supergene pro-
related to arsenic-rich mineralization cannot be calculated cesses are minor, based on logging as well as the absence of
without specific information on the mineralogical form of the samples with high As/Cu ratios. Below 3,250 m a.s.l., enargite
gold. The gold fraction above the enargite line on Figure 9B veins are only subordinate and the density of stage 1 stock-
correlates with the concentration of Cu, as shown by Figure work veining is well correlated with high Cu grades.
10. Because these samples are also located on a trend with The data from the upper zone, where drill sampling is least
an Au/Cu ratio of 0.1 ppm/wt % and are correlated with drill biased, allow an estimation of the minimum and maximum
core intervals that have a high density of stockwork veining, amount of copper hosted in enargite in this zone. About 16%
we conclude that up to 80% of the gold was introduced with of the Cu in the top zone lies within the compositional band
stage 1 mineralization. defined by the enargite As/Cu ratio (Fig. 9A). This fraction
The vertical distribution of vein density and metals con- most probably represents the minimum amount of Cu bound
tained in hypogene ore minerals has been plotted as a function to enargite in these elevations. Assuming that all As in the
of the present-day elevation (Fig. 11). The Cu, Au, As, and top part of the deposit is bound to enargite, the maximum
Mo grades all show maximum values at an elevation of about fraction of Cu related to enargite is calculated as 34%. This
3,400 ± 300 m, but this primarily reflects the depth interval proportion is unlikely to be realistic, and even fractions some-
with the best drill core coverage (Fig. 11). This sampling bias what lower than 16% copper in enargite are possible, because
prevents a quantitative resolution of grade variations with ore the upper zone contains an unquantified fraction of oxidized
mineralogy as a function of depth. Qualitatively, however, high samples that host secondary arsenate assemblages with par-
Cu grades and high stockwork vein density extend to greater tially leached copper.
depth than other ore elements and vein types. For the intermediate depth range of 3,400 to 3,250 m a.s.l.,
The four main vein types show a distinct variation as func- where less enargite veins have been logged, the uncertainty
tion of elevation, as shown by the color symbols in Figure 11 arising from the presence of secondary arsenates is consid-
(for clarity, only the sample intervals with vein densities 4 and erably reduced. For this zone, the fraction of Cu bound to
5 have been plotted and, for stage 4 enargite veins, also vein enargite is calculated to be 12 to 26%. In the deepest part of
density 3). Stage 4 veins appear to be relatively prominent the deposit (below about 3,250 m a.s.l.), the fraction of Cu
at shallower levels (Fig. 11A-C), whereas at greater depths, estimated to be hosted by enargite is only 5 to 17%.
stage 1 stockwork veins are most abundant (Fig. 11J-L). The The siting of gold at Altar in the three different depth
stage 3 tourmaline and stage 2 pyrite veins are moderately ranges has been estimated by summing up the intervals where
abundant over the entire vertical exposure of the deposit, but at least one enargite vein was mapped and, additionally, by
high vein densities do not correlate with high metal grades calculating the gold fraction of the intervals located within the
(Fig. 11D-I). uncertainty interval of the enargite line in Figure 9A. Approx-
Based on Cu grades, the elevation profile at Altar can be imately 10% of the Au above 3,400 m a.s.l. is located in inter-
divided into three distinct vertical zones. In the top zone vals where enargite veins are logged. For the middle and deep
(above 3,400 m a.s.l.), enargite veins probably dominate the zone, this fraction of Au decreases to 9% and 3%, respectively.
2
4.1 ppm Au
1.6
1.2
Au (ppm)
0.8
2 m scale exploration assay
2 m scale: stockwork veins density 5
0.4 u
2 m scale: stockwork veins density 4
wt.% C
m Au / Small-scale assay: stockwork veins
0.14 pp Small-scale assay: quartz-pyrite veins
Small-scale assay: enargite veins
0
0 0.5 1 1.5 2 2.5 3 4 6 8 10 12 14 16 18 20 22 24
Cu (wt.%)
Fig. 10. Correlation between Au and Cu grades and stockwork vein densities 4 and 5, based on bulk assay data (2-m drill
core intervals) and small-scale assay data of the main vein groups; 0.14 ppm Au/wt % Cu corresponds to the average resource
ratio in the deposit; note break in scale between 3 and 4% Cu; one sample at 4.1 ppm Au plots above the diagram.
1350 ZWAHLEN ET AL.
Stage 4: Enargite veins if semiquantitative, could be of considerable economic rele-

4000
A
4000
B
4000
C vance and warrants more rigorous geostatistical analysis prior
to mine development and metallurgical plant design.
Elevation (m)
3500 3500 3500

Individual vein analysis based on small-scale assays
3000 3000 3000 The distribution of Au has been further investigated by
small-scale assays and mineral analysis. Guided by the large-
2500 2500 2500 scale logging and assay data for 2-m intervals, a small number
0 1 2 3 0 2000 0 0.5 1
Cu (wt.%) Au (ppm)
of representative vein-scale samples (5–20 cm of half-core)
As (ppm)
were analyzed to estimate the amount of Au associated with
Stage 3: Tourmaline veins the different vein types. Individual veins along with their sur-
4000 4000 4000 rounding alteration halos were selected. Samples dominated
D E F
by stage 1 stockwork veins or stage 2 quartz-pyrite veins were
Elevation (m)
3500 3500 3500

selected from intervals that have high Au but low As grades in
order to remove the influence of enargite veins. Stage 4 enar-
3000 3000 3000
gite veins were sampled from intervals that have bulk As/Cu
concentrations close to enargite stoichiometry, and the results
confirm that 90% of Cu in such intervals was hosted by enar-
2500 2500 2500
0 1 2 3 0 2000 0 0.5 1 gite. Two samples of disseminated Cu in potassically altered
Cu (wt.%) As (ppm) Au (ppm) rocks (stage 1) that lacked veins were also analyzed to see
whether host-rock mineralization differs from stockwork vein
Stage 2: Quartz-pyrite veins
mineralization in terms of the Cu/Au ratio. Several samples
4000 4000 4000
G H I were taken within the same 2-m drill core interval in order to
evaluate whether different vein types have different Cu/Au
Elevation (m)
3500 3500 3500

ratios. Due to the high density of tourmaline veins observed in
many intervals that have low metal grades, we infer that stage
3000 3000 3000 3 veins did not contribute significantly to metal endowment.
In total, 16 stockwork veins (stage 1), 10 quartz-pyrite veins
2500
0 1 2 3
2500
0 2000
2500
0 0.5 1
(stage 2), and three enargite veins (stage 4) were selected. The
Cu (wt.%) As (ppm) Au (ppm) samples were ground to fine powder and two aliquots for each
sample were analyzed for major and trace elements. Aliquots
Stage 1: Stockwork veins of 30 g were taken from 100 to 200 g powder for gold fire assay.
4000
J
4000
K
4000
L In order to test for potential nugget effects due to discrete
gold grains, analysis of Au concentrations was performed on
Elevation (m)
3500 3500 3500 two powder aliquots from each sample. Another 0.5 g of each
sample was used for ICP-OES analysis. Geochemical analy-
3000 3000 3000 sis was performed by Acme Analytical Laboratories (Chile),
using the same procedures as for the exploration analyses.
2500 2500 2500 Figure 12 compares the Au and Cu concentrations of small-
0 1 2 3 0 2000 0 0.5 1
Cu (wt.%) Au (ppm)
scale samples (intensely colored thin bars, color indicating
As (ppm)
vein type) with those of the host interval from which they
Fig. 11. (A-L) Plots of the concentrations of Cu, As, and Au (vertical were taken (faintly colored broader bars, color indicating pre-
columns) as a function of present-day elevation. Diagrams are grouped in
four rows, which correspond to the four main vein types (stockwork veins,
dominant alteration type). Most of the 16 selectively sampled
quartz-pyrite veins, tourmaline veins, enargite veins). Element concentra- stage 1 stockwork veins contain higher Cu and Au concen-
tions plotted as circles indicate high vein density between 4 (smaller circles) trations than the average grades of the 2-m intervals from
and 5 (larger circles) of each of the four vein types. which they originate (Fig. 12), but the ratios of Au/Cu were
similar in both the small-scale sample and the correspond-
Alternatively, the gold fraction above 3,400 m within those ing larger interval (Fig. 10). This observation confirms the
intervals where all copper is thought to be present as enargite predominant contribution of the stockwork veins to the bulk
is 27%. This fraction decreases for the middle and deep zones Cu and Au grades. Cu and Au grades in small-scale samples
to 25% and 11.7%, respectively. dominated by later stage 2 quartz-pyrite veins are similar to
In summary, our first-order semiquantitative correla- the host interval (slightly lower or higher), indicating that the
tion analysis suggests that Cu and Au are correlated with metals had already been introduced by the pervasive earlier
each other and with the extent of stage 1 stockwork veining mineralization stage and were, at best, slightly redistributed
throughout the deeper parts of the deposit. However, in the by the stage 2 event. Even in intervals with pervasive feldspar-
top 400 m of the hypogene orebody, about one-third of the destructive overprint, the related intense stage 2 quartz-pyrite
sulfide-hosted Cu is likely to be bound to enargite, and is veining does not elevate Cu or Au above the concentration in
associated with a disproportionately large fraction of the gold the background interval, indicating that the rock was previ-
(Au/Cu ratios of 0.2 ppm/wt % compared to 0.14 ppm/wt % ously mineralized by stage 1 stockwork veins and the associ-
averaged for the entire Altar orebody). This conclusion, even ated potassic alteration.
213-1
200-1
2 4.1
215-1
200-2
1.5
214-1
Au (ppm)
Assay data of stockwork vein sample
204-2
203-1
1 204-1
210-1
204-3
Assay data of 2 meter interval
213-2
203-2
210-2
206-1
205-2
205-1
209-1
201-2
207-1
207-2
208-1
208-3
212-1
201-1
208-2
202-1
0.5
211-1
205-3
207-3
0
2
Cu (wt.%)
Stockwork vein Pervasive potassic alteration

8
Quartz-pyrite vein Potassic alteration with partial feldspar-destructive overprint
Enargite vein Pervasive overprint by feldspar-destructive alteration 24 17.5
10
12.6
Fig. 12. Results of small-scale assays of individual stockwork veins, quartz-pyrite veins, and enargite veins (thin bars in
blue, green, and orange, respectively) compared with the average Au and Cu grades of longer 2-m intervals from which
the small samples were taken. Comparison of the thin bars with the height of the broader blocks, light-colored according
to predominant alteration in the interval (from pervasive potassic alteration to partial and complete overprint by feldspar-
destructive alteration), shows whether individual veins of each type preferentially contribute to the bulk ore grade. Interval
200 = ALD-49 318 to 320 m; 201 = ALD-49 418 to 420 m; 202 = ALD-49 430 to 432 m; 203 = ALD-49 436 to 438 m; 204 =
ALD-49 442 to 444m; 205 = ALD-49 458 to 460 m; 206 = ALD-46 328 to 330 m; 207 = ALD-46 338 to 340 m; 208 = ALD-49
570 to 572 m; 209 = ALD-132 516 to 518 m; 210 = ALD-127 98 to 100 m; 211 = ALD-124 78 to 80 m; 212 = ALD-76 430 to
432 m; 213 = ALD-76 258 to 260 m; 214 = ALD-88 148 to 150 m; 215 = ALD-123 154 to 156 m.
Stage 4 enargite veins, by contrast, contain substantially Laser ablation-inductively coupled plasma-mass spectrom-
more Cu and Au than stage 1 stockwork veins (Fig. 12), etry used a beam-homogenized 193-nm ArF excimer laser
clearly indicating net addition of both metals. Data of a stage system connected to a quadrupole ICP-MS (Perkin Elmer
4 vein (213-1) from the same 2-m drill core interval as a stage Elan 6100 DRC). The laser was operated at a 10-Hz pulse
1 stockwork vein (213-2) show much higher Cu and Au, but in rate with an energy density of 5 to 6 J cm–2. The ablation diam-
comparable element proportion. eter was 90 µm, except for a few small pyrite grains where a
diameter of 40 µm was used. Approximately 0.1 to 0.2 µm/
Grain-Scale Distribution of Gold pulse was ablated from pyrite and chalcopyrite. Trace ele-
We selected 17 of the small-scale assay samples of stage ment abundance was measured using the following isotopes:
1, 2, and 4 veins for further detailed petrography in order 32S, 57Fe, 59Co, 62Ni, 65Cu, 66Zn, 75As, 77Se, 95Mo, 101Ru, 103Rh,
to determine the vein mineralogy and paragenesis with a 105Pd, 108Pd, 109Ag, 121Sb, 125Te, 189Os, 193Ir, 197Au, 202Hg, 208Pb,
focus on further identifying the siting of gold in the ore and 209Bi. Measurement of 40 s of gas blank was followed by
veins. Reflected-light microscopy on polished thick sections 60 s of sample ablation. For calibration, the sulfide standard
was assisted by scanning electron microscopy (SEM) imag- Laflamme Po724-B3 (Sylvester et al., 2005) was used as an
ing (with an energy dispersive spectroscopy [EDS] detector) external reference material for Au and the platinum group
and followed by LA-ICP-MS spot analysis of pyrite and other elements, and NIST 610 for all other elements. The latter
sulfides, including (sub)microscopic inclusions contained in is a nonmatrix matched silicate standard, but the difference
them. between the gold concentrations calculated with the two stan-
dards was found to be within 1σ analytical error. The dwell
Analytical methods time was set to 0.1 s for all elements except Au and As, where
Scanning electron microscopy imaging and semiquantita- 0.3 and 0.2 s were used. The major element composition of all
tive analysis of mineral composition was performed with a pyrite grains was analyzed with the electron microprobe prior
Jeol JSM-6390LA instrument operated with an acceleration to the LA-ICP-MS measurements and found to contain less
voltage of 15 kV and a working distance of 10 mm. Quantita- than 0.02 wt % As according to the calculated detection limit
tive analyses of pyrite and chalcopyrite were performed with a (LOD) of the microprobe. Therefore, all LA-ICP-MS data
JEOL JXA-8200 electron probe microanalyzer (EPMA; 20 kV were internally standardized to the stoichiometric concentra-
acceleration voltage; 11.8 nA beam current for S and Fe and tions of iron in pyrite (FeS2) and Cu in chalcopyrite (CuFeS2).
12.2 nA beam current for As and Cu; B007.PYR standard for
Fe and S, B012.CHP standard for Cu, and B025.ASP stan- Microtextural observations
dard for As) to achieve lower detection limits, notably for Bornite and chalcopyrite precipitated in stage 1 stockwork
minor elements in pyrite. veins, but bornite is only recorded as inclusions within early
1352 ZWAHLEN ET AL.
pyrite, which locally occurs in association with chalcopyrite by chalcopyrite (Fig. 13D). Similar textural relationships
(Fig. 13A, B). This early pyrite contains very fine grained between pyrite and chalcopyrite are also visible at the mar-
(<1-µm) inclusions of Se-Ag-Te and Ag-Cd–bearing phases, gin of a later pyrite generation, emphasizing coprecipitation
which could be detected by qualitative SEM-EDS spectra. of pyrite and chalcopyrite in several stages (Fig. 13E). The
Early pyrite also occurs as relict crystals in stage 2 quartz- latest pyrite crystals related to stage 2 have sharp boundaries
pyrite veins, indicating reopening of stage 1 fractures. Rutile and show no subsequent replacement by chalcopyrite (Fig.
has replaced titanomagnetite in altered wall rocks (Fig. 13F). Apart from these associations, different chalcopyrite
13C). Early pyrite has been corroded and partially replaced generations do not have other diagnostic textural differences.
A B C
py
py
py
cp
cp
ru
100 µm 100 µm 100 µm
D E F
py
py
py
cp
qz
cp
100 µm 50 µm 500 µm
qz G I
H
cp
py
en Au
en cp
py
200 µm 400 µm 10 µm
J K cp L
cp
py qz
Au en
Au
cp Au py
25 µm 25 µm 200 µm
Fig. 13. Photomicrographs in reflected light showing representative textures of ore mineral assemblages. (A) Stage 1 early
pyrite with bornite and chalcopyrite inclusions. (B) Stage 1 stockwork vein with coarse-grained pyrite and some chalcopyrite.
(C) Titanomagnetite decomposed into rutile and pyrite. (D) Stage 1 pyrite replaced by chalcopyrite. (E) Stage 2 quartz-pyrite
vein with late pyrite partially replaced by late chalcopyrite. (F) Stage 2 pyrite vein with sharp borders. (G) Stage 4 enargite-
pyrite vein. (H) Stage 4 enargite with relicts of replaced chalcopyrite from an earlier vein stage. (I) Stage 1 resorbed pyrite
with gold grains. (J) Gold grains in stage 1 stockwork vein next to pyrite. (K) Gold grains in stockwork vein next to chalcopy-
rite. (L) Large gold grain associated with coarse-grained stage 4 enargite. Abbreviations: cp = chalcopyrite, en = enargite, py
= pyrite, qz = quartz, ru = rutile.
Late-stage pyrite crystals in stage 2 quartz-pyrite veins are dis- A representative time-resolved LA-ICP-MS signal of inclu-
tinct; they do not overgrow earlier sulfide minerals and are sion-free chalcopyrite from the stockwork vein sample 204-2 is
characterized by coarse-grained texture and the absence of shown in Figure 14A, recording stable major element concen-
silicate or sulfide inclusions (Fig. 13F). trations and constant but low trace element concentrations.
Enargite is rarely present as disseminations and, more com- Concentrations of Bi, Pb, Ag, and Te in this chalcopyrite are
monly, as crystals filling discrete veins, associated with pyrite 0.4 ppm, 1.7 ppm, 3 ppm, and 0.5 ppm, respectively. The Au
and minor amounts of gypsum (pseudomorphs after anhy- concentrations are below the limit of detection of 0.02 ppm.
drite), native sulfur, and quartz (Fig. 13G). Pyrite in enargite Some optically homogeneous pyrite grains also show constant
veins is virtually devoid of inclusions and displays sharp grain ablation signals, as exemplified by Figure 14B, indicating a
boundaries with the adjacent enargite. Rarely, small pyrite low concentration of 150 ppm Cu, whereas Au, Bi, Te, and Ag
and chalcopyrite inclusions within enargite occur, which indi- are all below their LODs of 0.01 ppm, 0.01 ppm, 0.27 ppm,
cate that this Cu-As sulfide formed somewhat later in the and 0.02 ppm, respectively. Other samples contain apparently
hydrothermal evolution (Fig. 13G, H). constant, detectable concentrations of trace elements, and
Additional vein minerals, locally present in low abundance, Table 1 reports the concentration ranges found in apparently
include small grains of galena, tennantite, and monazite, homogeneous pyrites and chalcopyrites. The highest value
which commonly occur adjacent to quartz and pyrite. They of Au that is possibly present in homogeneous solid solution
form micrometer-sized grains and were only detected with (0.6 ppm) occurs in pyrite from a quartz-pyrite vein (sample
SEM-EDS. Galena grains were rarely observed in stage 1 204-3).
stockwork and stage 2 quartz-pyrite veins, whereas tennan- Figure 14C to G shows transient signals from pyrites con-
tite and monazite were exclusively detected in stage 1 stock- taining visible (up to 30 µm in diameter) or commonly invis-
work veins. Native gold was found in four samples. Gold ible inclusions, mostly occurring as texturally early pyrite in
grains are 2 to 150 µm in diameter and contain approximately stockwork veins. These inclusions contain Cu as a major com-
10 wt % silver (Fig. 13I-L), based on semiquantitative SEM- ponent and additionally have highly elevated concentrations
EDS measurements. In two stage 1 stockwork vein samples, of Bi, Ag, Te, Pb, and Au, which not only exceed those in the
gold grains were observed adjacent to inclusion-rich pyrite host pyrite, but are also much higher than their concentra-
between chalcopyrite and quartz crystals (Fig. 13J). In one tions in all homogeneous chalcopyrite grains occurring outside
sample, gold is present as inclusions within chalcopyrite adja- pyrite. During laser ablation, these composite inclusions com-
cent to a quartz grain (Fig. 13K). In three stage 1 stockwork monly account for only a small fraction of the entire volume of
veins, the gold grains are associated with disseminated sul- the ablation crater (probably at most 10% by mass in the inte-
fide grains in altered wall rock. Several gold grains 2 µm in grated signal), but they account for a significant bulk concen-
diameter were found within a partially replaced pyrite crystal tration of trace elements in these pyrites. The composition of
associated with stage 1 stockwork veins (Fig. 13I). The larg- the copper-rich inclusion phase in pyrite is not constant in all
est native gold grains have been observed within an enargite inclusions, or the elements are heterogeneously distributed
vein (Fig. 13L). These textural relationships leave no doubt within the inclusions, as indicated by imperfect covariation
that gold precipitated during the enargite vein stage of the of Cu with trace elements even within one pyrite host (Fig.
hydrothermal system. Given the particularly large size of this 14C-G). The composition of the originally trapped inclusion
enargite-hosted gold (Fig. 13L) and the selective enrichment phase(s) therefore cannot be determined reliably, but we can
of Au in other veins (Fig. 12), metallurgical processing of gold approximately quantify the average contents of trace elements
from the upper part of the orebody will need to account for in these inclusions relative to Cu, assuming a stoichiometry
the enargite-rich zone. for the dominant Cu-Fe sulfide phase. For example, if we
assume that the well-resolved inclusion recorded in the abla-
LA-ICP-MS analysis of trace element distribution in pyrite tion period 79 to 105 s of Figure 14C approximates the major
In addition to free-milling native gold, an unknown but element stoichiometry of chalcopyrite, then the integrated
probably significant fraction of invisible gold is located as element proportions relative to Cu translate into concentra-
microscopic or even submicroscopic particles within sulfide tions of 4 ppm Au, 380 ppm Pb, 4,300 ppm Bi, 92 ppm Te,
minerals. A potential host phase for invisible gold is the pyrite and 360 ppm Ag in the inclusion phase. If the phase were
itself, in particular when it contains high As concentrations closer to the bulk composition of bornite, the minor element
(Reich et al., 2005; Cook et al., 2009). To assess invisible gold concentrations would be about 1.8 times higher.
in the Altar deposit, we have explored its distribution together Figure 14D shows the more common situation of an
with other trace elements in 56 pyrite and chalcopyrite crys- LA-ICP-MS signal of a pyrite containing numerous small
tals from 12 samples. and poorly resolved inclusions (e.g., Fig. 13A, B, J), which
Table 1. Results of LA-ICP-MS Measurements of Trace Element Concentrations in Pyrite and Chalcopyrite
Mineral Au (ppm) Cu (ppm) Ag (ppm) As (ppm) Bi (ppm) Pb (ppm) Co (ppm) Ni (ppm) Zn (ppm) Se (ppm) Te (ppm)
Pyrite 0.1–0.6 80–3,800 0.2–4.3 0.1–1.3 0.3–1.3 0.3–11 5–20 5–30 1.5–4 80–300 1–4
Chalcopyrite 0–0.3 346,000 3.5–27 0.2–1 0.5–6 2.4–3.2 1.2–106 32–53 13–43 134–148 0.6–6.4
Note: Trace element concentrations of apparently homogeneous signals or signal intervals, corrected for instrument background and quantified based on
stoichiometric composition of Fe in pyrite and Cu in chalcopyrite
1354 ZWAHLEN ET AL.
A B C D
10 8
Fe
Fe
10 7 Cu
S
S
10 6
Cu
Counts
Bi
Pb
10 4 Pb
Ag Ag
10 3
Te low -
Pb, Te low - omitted omitted
Bi
10 2 Au
Te Te
Au
10 1
40 60 80 100 40 60 100 40 60 80 100 60 80 100
Time (s)
E F G Pyrite
Early Cu-Fe-S phase
10 8 Chalcopyrite
Fe Iron
Fe S Sulfur
10 7 Cu Copper
Bi Bismuth
S Pb Lead
10 6 Cu Ag Silver
Au Gold
Te Tellurium
Counts
Bi
10 4
Ag
Pb
Pb, Te low - omitted Au
10 3
Te
10 2
10 1
40 50 60 70 80 90 40 80 100 40 60 80
Time (s)
Fig. 14. Typical LA-ICP-MS signals of sulfides, recording layer-by-layer analysis of host mineral and contained inclusion
phases, intersected by an advancing large-diameter (90 µm) ablation crater. Initially high Pb is a local surface contamination
of the polished section. (A) Homogeneous chalcopyrite. (B) Homogeneous pyrite. (C) Pyrite with two large inclusions of an
early Cu-Fe-S mineral with high Bi, Pb, and Ag but little Au or Te. (D) Pyrite and early Cu-Fe-S phase with variable contents
of trace elements including Au, indicating that the inclusions are now heterogeneous phase mixtures of tiny mineral grains.
(E) Pyrite and deep inclusion of early Cu-Fe-S phase with high Au, Te. (F) Pyrite and early Cu-Fe-S phase. (G) Ablating from
pyrite into enclosed chalcopyrite, with minor elements including Au concentrated at the transition, probably as micrograins
along internal grain boundary. The dominant mineral type in the transient signals is schematically indicated with colored bars
above each signal.
commonly have similar contents of Bi, Pb, and Ag relative to assume that it approximates chalcopyrite stoichiometry. Fig-
Cu, but significantly higher proportions of Au and Te. Inte- ure 14E shows the signal of an optically homogeneous pyrite
grating all trace elements through this pyrite sample yields that contains a small fraction of tiny Cu-rich inclusions in the
a concentration of 110 ppm Au, 460 ppm Pb, 2,080 ppm Bi, first part of the signal, as indicated by a somewhat elevated
160 ppm Te, and 370 ppm Ag in the inclusion phase, if we and more irregular Cu signal compared to the homogeneous
pyrite of Figure 14B. After 85 s the concentrations of Cu, Bi, Cu Cu

Pb, Te, Au, and Ag increase simultaneously, which is readily Ag-rich
explained by one larger Cu-rich inclusion or an increase in Cu-Fe sulfide
the proportion of microscopic inclusions compared to the host

pyrite. Figure 14F shows another inclusion-rich pyrite. The Bi-rich and
Te-poor phase
copper concentration of this grain is of the same magnitude (low-Ag-Cu)
Bi-telluride
as the pyrite shown in Figure 14C, but the variation in Cu
concentration is much smaller than those of other trace ele- Gold A B
Ag*1000 Bi*1000 Te*1000
ments like, for example, Bi. Figure 14G shows an LA-ICP-MS Au*10000
Ag
signal where pyrite was ablated at the surface before a large Cu
inclusion was intersected at greater depth, occupying the full Ag-rich

area of the ablation crater from 65 s to the end of ablation at Cu-Fe sulfide
85 s. Comparing the Cu, Fe, and S intensities of this interval

with those in Figure 14A confirms that this large inclusion is Bi-rich and
Te-poor phase
indeed close to chalcopyrite in composition. Note, however, Bi-rich and
Te-poor phase
that at the transition between host pyrite and this inclusion,
C D
elevated concentrations of gold and other trace elements (low-Ag-Cu)
Bi Au Bi*1000 Ag*1000
Bi-telluride
were observed, indicating that these elements occur as sepa-
Ag Ag
rate microscopic grains near the interface between the host
pyrite and the Cu-Fe sulfide inclusion. Bi-rich and
Te-poor phase
Similar to a recent study by Gregory et al. (2013), our high-
resolution LA-ICP-MS signals reveal information about the
internal heterogeneity of the inclusions which may result (low-Ag-Cu)
either from separate phases or from cooling and internal reor- Bi-telluride Gold
ganization. The molar trace element proportions of the early Pb-phase E F
Cu-Fe-S inclusion phases derived from LA-ICP-MS data are Pb Au Te Au
further explored in Figure 15, where each point was obtained
by integrating and quantifying the more or less distinct inclu-
sion peaks in all transient signals. Although no pure analysis Integrated signals of inclusion-rich pyrite
of a single inclusion phase could be obtained due to the small Single inclusions of early Cu-Fe-S phase in pyrite
Chalcopyrite grains
grain size, correlations in the ternary diagrams suggest that Native gold grains
Deposit average of Cu, Au and Ag
the compositional variation requires at least five end-member Deposit average metal ratios
phases. These end-members are native gold with low silver Fig. 15. (A-F) Molar element concentrations of inclusion phases trapped
content, one Cu-Fe-Ag sulfide phase, one Bi-rich phase, one in early pyrite, obtained by integrating and partly deconvoluting distinct
Bi telluride phase, and one Pb-rich phase. Based on the varia- peaks in transient LA-ICP-MS signals. Green symbols indicate deposit aver-
age, chalcopyrite data, and native gold grains.
tion in Bi/Te ratios, the Bi telluride phase could be tetrady-
mite (Bi2Te2S) or tsumoite (BiTe; Cook et al., 2007). The
Pb-rich phase is most likely galena. incorporated in a homogeneous Cu-Fe-S–rich solid solution
The analyzed arsenic content is low in all homogeneous phase (as suggested by Simon et al., 2000, and Kesler et al.,
pyrite grains (Table 1) as well as the inclusions, suggesting 2002), which was trapped in pyrite and subsequently formed
that incorporation of invisible Au into pyrite or Cu-Fe sul- the different mineral phases indicated in Figure 15 by inter-
fide was not controlled by As substitution, which is in contrast nal unmixing and reaction with the host pyrite.
to many sediment-hosted gold deposits. This interpretation
is supported by the high Au/As ratios of pyrite in stockwork Significance of Au-rich inclusions trapped in
veins (stage 1) and quartz-pyrite veins (stage 2), compared pyrite for bulk gold grades
to typical arsenian pyrites in gold deposits that host substan- The total amount of gold trapped in pyrite found by LA-
tial invisible gold (Richards and Kerrich, 1993; Reich et al., ICP-MS analysis can be compared to the small-scale assay
2005). The gold and arsenic concentrations found in homoge- data in order to obtain a rough estimate of the fraction of gold
neous pyrites from the Altar deposit are both lower than those hosted by pyrite. First, the fraction of pyrite in representative
reported for invisible Au in arsenious pyrite, whereas high samples was estimated from petrographic observations and
integrated gold contents from inclusion-rich pyrite are well converted into mass percentage (Table 2). Combined with the
above the limit of Au-As incorporation by coupled surface average gold concentration in selected pyrites determined by
processes, as observed in Carlin deposits (Reich et al., 2005). LA-ICP-MS, the Au concentration locked up in pyrite in each
This confirms that gold now occurs as elemental Au0 enclosed sample was calculated and then compared to the total gold
in the pyrite crystals at Altar. In summary, the LA-ICP-MS concentration obtained from the small-scale assay (Table 2).
signal shapes and our partial quantification of element ratios For our limited number of 13 samples where both types of
indicate that several distinct Au-, Te-, Bi-, and Pb-rich phases data were obtained, we calculate that an average fraction of
were either coprecipitated as small crystals together with chal- ~12% of the total contained gold is enclosed in pyrite. For
copyrite, prior to overgrowth by the early pyrite generation all samples of potassic alteration containing visibly inclusion
associated with stockwork veining, or that they were originally rich pyrite, this fraction is higher, at 22% on average. Samples
1356 ZWAHLEN ET AL.
Table 2. Estimated Fraction of Refractory Gold Hosted by Pyrite in Selected Small-Scale Core Samples
Sample Gold in py Modal py in Gold concentration Gold locked in Gold fraction

no.1 Description1 (vein type) (ppm)2 sample (wt %)3 in rock (ppm)4 pyrite (ppm in rock)5 in pyrite (%)6
Potassic alteration
204-1 Stockwork vn (1A) 1.01 18 0.80 0.18 22.7
204-2 Stockwork vn (1A) 0.25 4 0.92 0.01 1.1
205-2 Stockwork vn (1A) 0.89 2 0.35 0.02 5.1
206-1 Stockwork vn (1A) 0.81 2 0.36 0.02 4.5
204-3 Quartz-pyrite vn (2C) 1.19 35 0.46 0.42 90.5
205-1 Quartz-pyrite vn (2B) 0.10 20 0.29 0.02 6.9
Potassic alteration with partial

feldspar-destructive overprint
203-1 Stockwork vn (1A) 1.51 4 0.92 0.06 6.6
200-1 Stockwork vn (1A) 0.11 20 1.78 0.02 1.2
200-2 Quartz-pyrite vn (2B) 0.13 35 0.93 0.05 4.9
203-2 Quartz-pyrite vn (2A) 0.46 10 0.28 0.05 16.4
Pervasive overprint by feldspar-

destructive alteration
213-2 Stockwork vn (1A) 0.06 4 0.42 0.002 0.6
213-1 Enargite vn (4), clean coarse pyrite 0.04 9 4.13 0.004 0.1
215-1 Enargite vn (4), clean coarse pyrite 0.005 51 1.36 0.003 0.2
1 Small-scale assay sample, dominant vein (vn) and alteration type

2 Gold concentration in pyrite (py) obtained by complete integration of LA-ICP-MS signals from three to six pyrite grains in sample, referenced to stoi-
chiometric pyrite
3 Visually estimated volume fraction of pyrite, chalcopyrite, enargite, and silicates in sample, recalculated to wt % assuming pyrite density of 5 g cm–3,
chalcopyrite density of 4.2 g cm–3, enargite density of 4.4 g cm–3, and silicate rock density of 2.6 g cm–3
4 Commercial bulk assay of small-scale sample
5 Gold in pyrite × wt % pyrite in sample
6 As percentage fraction of total gold concentration in rock
containing no inclusion-rich pyrite have a smaller gold fraction combining semiquantitative vein density logging obtained
of less than 5% hosted by pyrite. From these observations we during exploration with standard assay data and smaller-scale
can conclude that most of the gold is located in the matrix— bulk assays and mineral microanalyses. Results indicate that
most likely as small native gold particles, even though these most of the moderately small but economically significant
are rarely detected by petrographic observation—but that a gold tenor of this deposit (bulk Au/Cu ratio: 1.4 × 10–5 by
significant fraction of gold in the deposit could be locked up as mass) was precipitated together with copper during the first
refractory inclusions in pyrite. Our observations closely match stockwork veining stage. About 10 to 20% of the gold is esti-
some of the conclusions of a recent QEMSCAN study by mated to be locked into early pyrite as small Cu-(Fe) sulfide
Gregory et al. (2013), who also emphasized the potential met- inclusions that are enriched in other chalcophile elements
allurgical importance of the gold fraction enclosed in pyrite. and Te. Later vein stages show no evidence of significant gold
addition, but some stage 1 gold may have been remobilized
Summary and Conclusions during the later mineralization stages. Stage 4 enargite veins,
The Altar porphyry deposit represents a major copper which are particularly abundant in the upper parts of the
resource with minor but significant gold and molybdenum deposit, may have contributed additional Cu and Au in similar
credits, related to a late Miocene dacite porphyry stock. proportion as in the potassic stage. We estimate that between
Mineralization and alteration extend into intermediate to fel- 11 and 26% of the total copper contained in the orebody is
sic volcanic rocks and, possibly, intrusive andesites of early hosted in stage 4 enargite veins, with lower values in the zone
Miocene age. Several porphyry fingers as well as an igneous of supergene enrichment but possibly higher proportions in
breccia are mapped on the surface and in drill core, but this the upper parts of the hypogene ore zone.
study focuses on the Central Porphyry that hosts the main The distribution of Cu and Au ratios at Altar is primarily
explored orebody. Early potassic alteration was overprinted controlled by stage 1 stockwork mineralization and potassic
by feldspar-destructive alteration at the periphery of the alteration. Although precipitation of an early Au-rich Cu-Fe
deposit. Advanced argillic alteration is best developed at high sulfide phase may have contributed to a correlated distribu-
elevations and as local halos around enargite veins. Four main tion of the two ore metals (Kesler et al., 2002), the overall
vein stages were distinguished, which are (1) stockwork veins, Au/Cu ratio in the deposit was probably determined by the
(2) quartz-pyrite veins, (3) tourmaline veins, and (4) enargite composition and source of the mineralizing magmas. Region-
veins, the latter related to a high sulfidation epithermal min- ally, the somewhat higher Au grades and lower molybdenum
eralization style. content at Altar, compared to the nearby porphyry deposits
We have attempted to estimate the proportion of Cu and of Los Pelambres, El Pachon, Rio Blanco, and Los Bronces
As sited in different vein generations and host minerals by and the even more Au poor giant porphyry-Cu-Mo deposit
of El Teniente in Chile, might relate to increasing distance Gustafson, L.B., and Hunt, J.P., 1975, Porphyry copper deposit at El Salva-
of the deposit from the trench of the subduction zone, as dor, Chile: Economic Geology, v. 70, p. 857–912.
Halter, W.E., Pettke, T., and Heinrich, C.A., 2002, The origin of Cu/Au ratios
observed elsewhere. Farther east toward the stable continent, in porphyry-type ore deposits: Science, v. 296, p. 1844–1846.
porphyry-epithermal ore systems, including La Famatina and Halter, W.E., Heinrich, C.A., and Pettke, T., 2005, Magma evolution and the
Bajo de la Alumbrera, contain even higher bulk Au/Cu ratios formation of porphyry Cu-Au ore fluids: Evidence from silicate and sulfide
in ores related to potassic alteration, and fluid and melt com- melt inclusions: Mineralium Deposita, v. 39, p. 845–863.
positions clearly indicate a strong magmatic source control on Heredia, N., Fernandez, L.R.R., Gallastegui, G., Busquets, P., and Colombo,
F., 2002, Geological setting of the Argentine Frontal Cordillera in the flat-
metal endowment, probably including interaction of mafic slab segment (30°00 '-31°30 ' S latitude): Journal of South American Earth
mantle and felsic crustal melts (Halter et al., 2005; Sillitoe Sciences, v. 15, p. 77–99.
and Perelló, 2005). Hollings, P., Cooke, D., and Clark, A., 2005, Regional geochemistry of Ter-
tiary igneous rocks in central Chile: Implications for the geodynamic envi-
Acknowledgments ronment of giant porphyry copper and epithermal gold mineralization:
Economic Geology, v. 100, p. 887–904.
Jeff Toohey and Peregrine Minerals Ltd. are thanked for Instituto de Investigaciones Geológicas, 1978, Carta geológica de Chile
granting access to the Altar deposit, and for their invaluable 1:250000: Santiago de Chile, Instituto Geografico Militar.
help and support during fieldwork. They generously made drill Kay, S.M., and Mpodozis, C., 2002, Magmatism as a probe to the Neogene
core and proprietary information available for publication, shallowing of the Nazca plate beneath the modern Chilean flat-slab: Journal
which is greatly appreciated. We are also grateful to Florencia of South American Earth Sciences, v. 15, p. 39–57.
Kay, S.M., Godoy, E., and Kurtz, A., 2005, Episodic arc migration, crustal
Márquez-Zavalía and Miguel Galliski for logistic support and thickening, subduction erosion, and magmatism in the south-central Andes:
discussion in the early stages of this research project. Markus Geological Society of America Bulletin, v. 117, p. 67–88.
Wälle is thanked for assistance with the LA-ICP-MS analysis. Kesler, S.E., Chryssoulis, S.L., and Simon, G., 2002, Gold in porphyry
Thoughtful reviews by Melissa Gregory, David Cooke, and an copper deposits: Its abundance and fate: Ore Geology Reviews, v. 21, p.
anonymous reviewer greatly helped to clarify the manuscript. 103–124.
Landtwing, M.R., Furrer, C., Redmond, P.B., Pettke, T., Guillong, M., and
Heinrich, C.A., 2010, The Bingham Canyon porphyry Cu-Mo-Au deposit.
REFERENCES
III. Zoned copper-gold ore deposition by magmatic vapor expansion: Eco-
Almandoz, G., Zulliger, G., and Márquez-Zavalía, F., 2005, Altar: Mineral- nomic Geology, v. 105, p. 91–118.
izacíon de alta sulfuración vinculada a un systema porfido cuprifero, San Large, R.R., Danyushevsky, L., Hollit, C., Maslennikov, V., Meffre, S., Gil-
Juan, Argentina: Asociación geológica Argentina, XVI Congreso Geológico bert, S., Bull, S., Scott, R., Emsbo, P., Thomas, H., Singh, B., and Foster,
Argentino, La Plata, Buenos Aires, Argentina, 2005, Proceedings, Actas 2, J., 2009, Gold and trace element zonation in pyrite using a laser imaging
p. 369–376. technique: Implications for the timing of gold in orogenic and Carlin-style
Arif, J., and Barker, T., 2004, Gold paragenesis and chemistry at Batu Hijau, sediment-hosted deposits: Economic Geology, v. 104, p. 635–668.
Indonesia: Implications for gold-rich porphyry copper deposits: Minera- Lattanzi, P., Da Pelo, S., Muso, E., Atzei, D., Elsener, B., Fantauzzi, M., and
lium Deposita, v. 39, p. 523–535. Rossi, A., 2008, Enargite oxidation: A review: Earth-Science Reviews, v.
Ballard, J.R., Palin, J.M., and Campbell, I.H., 2002, Relative oxidation states 86, p. 62–88.
of magmas inferred from Ce(IV)/Ce(III) in zircon: Application to porphyry Maydagan, L., Franchini, M., Chiaradia, M., Pons, J., Impiccini, A., Toohey,
copper deposits of northern Chile: Contributions to Mineralogy and Petrol- J., and Rey, R., 2011, Petrology of the Miocene igneous rocks in the Altar
ogy, v. 144, p. 347–364. region, main Cordillera of San Juan, Argentina. A geodynamic model within
Cook, N.J., and Chryssoulis, S., 1990, Concentration of invisible gold in the the context of the Andean flat-slab segment and metallogenesis: Journal of
common sulfides: Canadian Mineralogist, v. 28, p. 1–16.
South American Earth Sciences, v. 32, p. 30–48.
Cook, N.J., Ciobanu, C.L., Wagner, T., and Stanley, C., 2007, Minerals of
Maydagan, L., Franchini, M., Chiaradia, M., Dilles J., and Rey, R., 2014,
the system Bi-Te-Se-S related to the tetradymite archetype: Review of
Intrusion history of the Altar porphyry Cu-(Mo-Au) deposit (Argentina):
classification and compositional variation: Canadian Mineralogist, v. 45, p.
A complex magmatic-hydrothermal system with evidence of recharge pro-
665–708.
cesses: Economic Geology, in press.
Cook, N.J., Ciobanu, C.L., and Mao, J., 2009, Textural control on gold dis-
Murakami, H., Seo, J.H., and Heinrich, C.A., 2010, The relation between
tribution in As-free pyrite from the Dongping, Huangtuliang and Hougou
gold deposits, North China craton (Hebei Province, China): Chemical Cu/Au ratio and formation depth of porphyry-style Cu-Au ± Mo deposits:
Geology, v. 264, p. 101–121. Mineralium Deposita, v. 45, p. 11–21.
Cook, N.J., Ciobanu, C.L., Danyushevsky, L.V., and Gilbert, S., 2011, Minor Musu, E., Da Pelo, S., Lorrai, M., and Lattanzi, P., 2007a, Variations and
and trace elements in bornite and associated Cu-(Fe)-sulfides: A LA-ICP- controls on the Cu/As ratios in waters at the Furtei gold mine, Sardinia:
MS study: Geochimica et Cosmochimica Acta, v. 75, p. 6473–6496. International Mine Water Association (IMWA) Symposium, Cagliari, Italy,
Cooke, D.R., and Hollings, P., 2005, Giant porphyry deposits: Characteristics, 27–31 May, 2007, Proceedings, p. 441–446.
distribution, and tectonic controls: Economic Geology, v. 100, p. 801–818. Musu, E., Da Pelo, S., Lattanzi, P., and Lorrai, M., 2007b, Secondary min-
Deditius, A.P., Utsunomiya, S., Renock, D., Ewing, R.C., Ramana, C.V., eralogy at Furtei, Sardinia, Italy: Controls on toxic elements mobility, in
Becker, U., and Kesler, S.E., 2008, A proposed new type of arsenian pyrite: Bullen, T.D., Wang, Y.X., eds., Water rock interaction—Proceedings of
Composition, nanostructure and geological significance: Geochimica et WRI-12: London, Taylor and Francis Ltd., p. 625–628.
Cosmochimica Acta, v. 72, p. 2919–2933. Mutschler, F.E., Ludington, S., and Bookstrom, A.A., 2010, Giant porphyry-
Deyell, C.L., and Hedenquist, J.W., 2011, Trace element geochemistry of related metal camps of the world—a database: USGS Open-File Report
enargite in the Mankayan district, Philippines: Economic Geology, v. 99-556, Online Version 1.0, http://pubs.usgs.gov/of/1999/of99-556/.
106, p. 1465–1478. Perez, D., D’Odorico Benites, P., and Godeas, M.C., 2010, Reconocimiento
Fleet, M.E., and Mumin, A.H., 1997, Gold-bearing arsenian pyrite and mar- de alteración hidrotermal con el sensor aster, en el curso medio del Rio
casite and arsenopyrite from Carlin trend gold deposits and laboratory syn- Santa Cruz (31°40'S), Provincia de San Juan: Revista de la Asociación
thesis: American Mineralogist, v. 82, p. 182–193. Geológica Argentina, v. 66, p. 625–635.
Frank, M.R., Simon, A.C., Pettke, T., Candela, P.A., and Piccoli, P.M., 2011, Ramos, V., 1988, The tectonics of the Central Andes, 30° to 33°S latitude:
Gold and copper partitioning in magmatic-hydrothermal systems at 800°C Geological Society of America, Special Paper 218, p. 31–54.
and 100 MPa: Geochimica et Cosmochimica Acta, v. 75, p. 2470–2482. Reich, M., Kesler, S.E., Utsunomiya, S., Palenik, C.S., Chryssoulis, S.L., and
Gregory, M.J., Lang, J.R., Gilbert, S., and Olson Hoal, K., 2013, Geomet- Ewing, R.C., 2005, Solubility of gold in arsenian pyrite: Geochimica et Cos-
allurgy of the Pebble porphyry copper-gold-molybdenum deposit, Alaska: mochimica Acta, v. 69, p. 2781–2796.
Implications for gold distribution and paragenesis: Economic Geology, Richards, J.P., 2003, Tectono-magmatic precursors for porphyry Cu-(Mo-Au)
v. 108, p. 463–482. deposit formation: Economic Geology, v. 98, p. 1515–1533.
1358 ZWAHLEN ET AL.
Richards, J.P., and Kerrich, R., 1993, The Porgera gold mine, Papua New Sylvester, P.J., Cabri, L.J., Tubrett, M.N., McMahon, G., Laflamme, J.H.G.,
Guinea: Magmatic hydrothermal to epithermal evolution of an alkalic-type and Peregoedova, A., 2005, Synthesis and evaluation of a fused pyrrhotite
precious-metal deposit: Economic Geology, v. 88, p. 1017–1052. standard reference material for platinum-group element and gold analysis
Rosenbaum, G., Giles, D., Saxon, M., Betts, P.G., Weinberg, R.F., and Duboz, by laser ablation-ICPMS, in Tormanen, T.O., Alapieti, T.T., eds., 10th Inter-
C., 2005, Subduction of the Nazca Ridge and the Inca Plateau: Insights into national Pt Symposium: Espoo, Finland, Geological Survey of Finland, p.
the formation of ore deposits in Peru: Earth and Planetary Science Letters, 16–20.
v. 239, p. 18–32. Ulrich, T., Günther, D., and Heinrich, C.A., 1999, Gold concentrations
Secretaria de Industria, Comercio y Minería, Subsecretaria de Minería, Ser- of magmatic brines and the metal budget of porphyry copper deposits:
vicio Geológico Minero Argentino, Instituto de Geologia y Recursos Mine- Nature, v. 399, p. 676–679.
rales, 1999, Mapa geológico y minero de la Republica Argentina 1:500000: von Quadt, A., Erni, M., Martinek, K., Moll, M., Peytcheva, I., and Hein-
Secretaria de Industria, Comercio y Minería, sheet 12. rich, C.A., 2011, Zircon crystallization and the lifetimes of ore-forming
Sillitoe, R.H., and Perelló, J., 2005, Andean copper province: Tectonomag- magmatic-hydrothermal systems: Geology, v. 39, p. 731–734.
matic settings, deposit types, metallogeny, exploration and discovery: Eco- Wagner, T., Klem, R., Wenzel, T., and Mattsson, B., 2007, Gold upgrading in
nomic Geology 100th Anniversary Volume, p. 845–890. metamorphosed massive sulfide ore deposits: Direct evidence from laser-
Simon, G., Kesler, S.E., Essene, E.J., and Chryssoulis, S.L., 2000, Gold in ablation-inductively coupled plasma-mass spectrometry analysis of invisible
porphyry copper deposits: Experimental determination of the distribution gold: Geology, v. 35, p. 775–778.
of gold in the Cu-Fe-S system at 400° to 700°C: Economic Geology, v. Yanez, G.A., Ranero, C.R., von Huene, R., and Diaz, J., 2001, Magnetic
95, p. 259–270. anomaly interpretation across the southern Central Andes (32°–34°S): The
Simpson, R.G., Nilsson, J., and Schlitt, W.J., 2011, Altar project, San Juan role of the Juan Fernandez Ridge in the late Tertiary evolution of the mar-
Province, Argentina: Peregrine Metals Ltd., Nl 43-101 Technical Report, gin: Journal of Geophysical Research: Solid Earth, v. 106, p. 6325–6345.
151 p.
View publication stats

Zwahlen 2014

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Zwahlen 2014

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Zwahlen 2014

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

The Porphyry Cu-(Mo-Au) Deposit at Altar (Argentina): Tracing Gold

Article in Economic Geology · May 2014

Carmen Zwahlen Thomas Wagner

SEE PROFILE SEE PROFILE

Element Mobility in Hydrothermal Systems View project

The user has requested enhancement of the downloaded file.

The Porphyry Cu-(Mo-Au) Deposit at Altar (Argentina):

Introduction by substitution into bornite and, to a minor extent, into chal-

111 110 108 94

131 128 86 98 124

93 121 123 84 127 122 130 Altar East

133 101 66 53 51 75 87 97 Late Miocene

2800m 2800m Central Porphyry

Potassic alteration Potassic alteration

A B following Almandoz et al., 2005), which consists of four stages

Vein stages Substages - vein mineralogy Potassic Feldspar-destructive Advanced argillic

1A Gray quartz with sulfides (py, cp, ±bn, ±anh)

1B Molybdenite - quartz / molybdenite - quartz with py and cp

1D White quartz with pyrite, chalcopyrite and feldspar halo

2A Quartz - pyrite (chalcopyrite) with silicic halo

2B Quartz - pyrite (chalcopyrite) with sericite - tourmaline halo

2E Hematite - magnetite with sulfides (± py, ±cp)

1cm 2mm 5mm 5mm

5mm 1cm 1cm 5mm

1cm 1cm 1cm

dominant ore minerals. In this analysis, assay data have been

copper at Altar. Many samples (6,967 out of 9,264) contain

Stage 4: Enargite veins if semiquantitative, could be of considerable economic rele-

3500 3500 3500

3500 3500 3500

3500 3500 3500

Assay data of stockwork vein sample

Stockwork vein Pervasive potassic alteration

100 µm 100 µm 100 µm

pyrite of Figure 14B. After 85 s the concentrations of Cu, Bi, Cu Cu

the proportion of microscopic inclusions compared to the host

inclusion was intersected at greater depth, occupying the full Ag-rich

85 s. Comparing the Cu, Fe, and S intensities of this interval

Sample Gold in py Modal py in Gold concentration Gold locked in Gold fraction

Potassic alteration with partial

Pervasive overprint by feldspar-

1 Small-scale assay sample, dominant vein (vn) and alteration type

View publication stats

You might also like