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

net/publication/344381220

Geochemical and Isotopic Signature of Pyrite as a Proxy for Fluid Source and
Evolution in the Candelaria-Punta del Cobre Iron Oxide Copper-Gold District,
Chile

Article  in  Economic Geology · September 2020

DOI: 10.5382/econgeo.4765

CITATIONS READS
0 231

4 authors, including:

Irene del Real Contreras Adam Simon

University of Chile University of Michigan
11 PUBLICATIONS   33 CITATIONS    133 PUBLICATIONS   2,220 CITATIONS   

SEE PROFILE SEE PROFILE

Martin Reich
University of Chile
164 PUBLICATIONS   3,337 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Geochemistry of cold springs and thermal fluids discharging in the Andes of Central Chile (32.5° and 36°S) View project

Magmatic and structural controls of the iron oxide copper gold Candelaria-Punta del Cobre district, Atacama region, Chile View project

All content following this page was uploaded by Irene del Real Contreras on 30 September 2020.

The user has requested enhancement of the downloaded file.

©2020 Society of Economic Geologists, Inc.
Economic Geology, v. XXX, no. XX, pp. X–X

Geochemical and Isotopic Signature of Pyrite as a Proxy for Fluid Source and Evolution
in the Candelaria-Punta del Cobre Iron Oxide Copper-Gold District, Chile
I. del Real,1,† J.F.H. Thompson,1 A. C. Simon,2 and M. Reich3,4
1Department of Earth and Atmospheric Sciences, Cornell University, Snee Hall, Ithaca, New York 14850
2Department of Earth and Environmental Sciences, University of Michigan, Ann Arbor, Michigan 48109
3Department of Geology and Andean Geothermal Center of Excellence (CEGA), Facultad de Ciencias Físicas y Matemáticas (FCFM),
Universidad de Chile, Plaza Ercilla 803, Santiago 8320000, Chile
4Millennium Nucleus for Metal Tracing Along Subduction, Facultad de Ciencias Físicas y Matemáticas (FCFM),
Universidad de Chile, Santiago 8320000, Chile

Abstract
Pyrite is ubiquitous in the world-class iron oxide copper-gold (IOCG) deposits of the Candelaria-Punta del
Cobre district, documented from early to late stages of mineralization and observed in deep and shallow levels
of mineralized bodies. Despite its abundance, the chemical and isotopic signature of pyrite from the Cande-
laria-Punta del Cobre district, and most IOCG deposits worldwide, remains poorly understood. We evaluated
in situ chemical and isotopic variations at the grain scale in a set of pyrite-bearing samples collected through-

T
out the district in order to characterize and further understand the nature of mineralization in this IOCG
system. Our multianalytical approach integrated synchrotron micro-X-ray fluorescence (µ-XRF) mapping of

F
pyrite grains with electron probe microanalysis and laser ablation-inductively coupled plasma-mass spectrom-
etry data, and sulfur isotope determinations using secondary ion mass spectrometry (SIMS) complemented
with bulk sulfur isotope analyses of coeval pyrite, chalcopyrite, and anhydrite. Synchrotron µ-XRF elemental

A
concentration maps of individual pyrite grains reveal a strong zonation of Co, Ni, As, and Se. The observed rela-
tionships between Ni and Se are interpreted to reflect changes in temperature and redox conditions during ore

R
formation and provide constraints on fluid evolution. Co and Ni concentrations and ratios suggest contributions
from magmas of mafic-intermediate composition. Pyrite chemical concentrations reflect potential stratigraphic
controls, where the sample from the upper part of the stratigraphy diverges from trends formed by the rest of

D
the sample set from lower stratigraphic levels. The SIMS δ34S values of pyrite (and chalcopyrite) range between
–2 up to 10‰, and bulk δ34S values of pyrite range between 4 up to 12‰. The majority of the δ34S analyses,
falling between –1 and 2‰, indicate a magmatic source for sulfur and, by inference, for the hydrothermal ore
fluid(s). Variation in the δ34S signature can be explained by changes in the redox conditions, fluid sources, and/
or the temperature of the hydrothermal fluid. The Se/S ratio combined with δ34S values in pyrite is consistent
with mixing between a magmatic-hydrothermal fluid and a fluid with a probable basinal signature. The results
of this study are consistent with the hydrothermal fluids responsible for mineralization in the Candelaria-Punta
del Cobre district being predominantly of magmatic origin, plausibly from mafic-intermediate magmas based
on the Ni-Co content in pyrite. External fluid incursion, potentially from a basinal sedimentary source, occurred
late in the evolution of the system, adding additional reduced sulfur as pyrite. There is no evidence to suggest
that the late fluid added significant Cu-Au mineralization, but this cannot be ruled out. Finally, the data reveal
that trace element ratios coupled with spatially resolved sulfur isotope data in pyrite are powerful proxies to
track the magmatic-hydrothermal evolution of IOCG systems and help constrain the source of their contained
metals.

Introduction et al., 2017; Simon et al., 2018), with Fe-rich and S-poor IOA
The source of hydrothermal fluids in iron oxide copper-gold mineralization representing the deeper roots of some Andean
(IOCG) deposits has been a controversial topic for several IOCG systems.
decades, since the discovery in the 1970s of the giant Precam- IOCG deposits are complex, relatively rare, and overall less
brian Olympic Dam deposit in Australia (Roberts and Hud- studied than well-understood systems such as porphyry cop-
son, 1983; Hitzman et al., 1992). Multiple interpretations and per deposits. The complexity in IOCG deposits may reflect
models have been proposed to explain the formation of these diverse tectonic settings, a range of host-rock and struc-
deposits involving different fluids, including oxidized saline tural environments, and the involvement of more than one
brines derived from evaporites (Barton and Johnson, 1996; hydrothermal fluid. The lack or scarce amount of quartz in
Hitzman, 2000), magmatic fluids derived from major intru- IOCG deposits hinders the use of fluid inclusions for deter-
sions (Marschik and Fontboté, 2001b; Pollard, 2006), and mining temperature, fluid composition and salinity, and the
combinations of both magmatic and basinal fluids (Williams potential source of hydrothermal fluids (Groves et al., 2010).
et al., 2005; Chiaradia et al., 2006). A connection to Kiruna- Even though alteration paragenesis has been well defined
type iron oxide apatite (IOA) deposits has also been proposed in some areas (e.g., the Great Bear district; Corriveau et al.,
(Sillitoe, 2003; Knipping et al., 2015a; Reich et al., 2016; Barra 2016), it can vary from deposit to deposit, with the presence
of iron oxides and Cu-rich sulfides being the only common
†Corresponding author: e-mail, [email protected] features in all IOCG deposits (Hitzman, 2000). The use of
ISSN 0361-0128; doi:10.5382/econgeo.4765; 25 p.
Digital appendices are available in the online Supplements section. 1 Submitted: July 19, 2019 / Accepted: March 17, 2020
2 DEL REAL ET AL.

mineral chemistry and isotopic systems provides alternative varying concentrations of trace elements in pyrite grains. Syn-
approaches for assessing the nature of the hydrothermal fluid chrotron µ-XRF was used for creating element concentration
and characterizing IOCG deposits (e.g., De Haller et al., imagery maps of the pyrite grains used for this study. LA-ICP-
2006; Rusk et al., 2010; Reich et al., 2016, among others) MS data were integrated with in situ sulfur isotope analyses
Most IOCG deposits contain pyrite (FeS2), ranging from an (δ34S) of pyrite and coexisting chalcopyrite using SIMS. Bulk
accessory phase in pyrite-poor systems (~<1%; e.g., Salobo, sulfur isotope data of pyrite and anhydrite determined by
Brazil, and Olympic Dam, Australia; Haynes et al., 1995; isotope ratio mass spectrometry (IRMS) were used to con-
Requia and Fontboté, 1999) to major amounts in pyrite-rich strain temperature variations of the hydrothermal fluid. We
systems (~>1%; e.g., Ernest Henry, Australia, and Cande- relate the chemical zonation of trace elements in pyrite with
laria, Chile; Marschik and Fontboté, 2001b; Williams et al., changes in the depth of formation, based on stratigraphic rela-
2005; Rusk et al., 2010). The presence of pyrite in most IOCG tionships and the mineral paragenesis in the district. We also
deposits provides an opportunity to compare their character correlate chemical variation in pyrite with variations of sulfur
and composition and potentially to use the resulting data to isotopes within the same pyrite grains. Our extensive data set
constrain hydrothermal fluid sources and evolution. Pyrite has provides constraints on fluid evolution, sources, temperature,
been successfully used as a proxy for characterizing hydro- and redox conditions at the Candelaria-Punta del Cobre dis-
thermal fluids in a large variety of mineral deposits, includ- trict with implications for the origin of the district and IOCG
ing Carlin-type gold, volcanogenic massive sulfide (VMS), systems.
sedimentary-hosted Cu/U, Archean to Mesozoic lode, and
epithermal gold deposits, among others (Cook and Chryssou- Geologic Background
lis, 1990; Hannington et al., 1999; Reich et al., 2005; Barker IOCG deposits in north-central Chile form part of the
et al., 2009; Cook et al., 2009; Large et al., 2009; Peterson and Andean IOA-IOCG belt, which extends from immediately
Mavrogenes, 2014; Gregory et al., 2015; Keith et al., 2016; north of Santiago to north of Antofagasta with the majority
Tanner et al., 2016; Román et al., 2019). Pyrite chemistry has of the deposits located between Tal Tal and Vallenar (Fig. 1).
been used as a record of fluid evolution in porphyry copper The belt continues north from the southern border of Peru
systems (Reich et al., 2013), and, in some cases, it has pro- to Lima for a total length in the Andes of ~2,000 km. Deposit
vided evidence for the transition from porphyry to epithermal ages range from ~90 to ~165 Ma, with El Espino being the
environments (Franchini et al., 2015). Pyrite chemistry has youngest at 88.4 ± 1.2 Ma (Lopez et al., 2014; del Real and
also been used in the Los Colorados IOA deposit (Chile) to Arriagada, 2015) and Montecristo and Julia the oldest at 164 ±
suggest a link with IOCG deposits (Reich et al., 2016). Finally, 11 and 159 ± 3 Ma (Boric et al., 1990; Espinoza et al., 1996).
pyrite chemistry has been evaluated to constrain the source of Within the belt, Mantoverde, Raúl Condestable, Mina Justa,
hydrothermal fluids in three IOCG deposits: Sossego, Brazil and Candelaria are major copper deposits, with Candelaria
(Monteiro et al., 2008), Ernest Henry, Australia (Rusk et al., being the most significant producer. Candelaria came into
2010), and more recently, the Marcona IOA and Mina Justa production in 1995 and current (September 2019) open-pit
IOCG deposits in Peru (Li et al., 2018). proven and probable reserves are 459 million tonnes (Mt) at
In addition to using the chemical composition of pyrite to 0.5% Cu (www.lundin.com). If all the mines and past produc-
constrain hydrothermal processes, the sulfur isotope compo- ers in the Candelaria-Punta del Cobre district are considered,
sition (δ34S) of pyrite can constrain the fluid source and, by the total endowment is approximately 13 Mt of contained Cu
inference, metal sources and ore-forming processes (Ohmoto, (del Real et al., 2018).
1972; Rye and Ohmoto, 1974; Ohmoto and Goldhaber, 1997). IOCG deposits in northern Chile are commonly spatially
This tool has been used on pyrite separates from IOCG depos- associated with, or hosted in, faults that form part of the
its where variable δ34S indicates a complex fluid source history Atacama fault system (Fig. 1; Arabasz, 1971; Grocott et al.,
(Marschik and Fontboté, 2001b; Benavides et al., 2007; De 1994; Espinoza et al., 1996; Grocott and Taylor, 2002). As with
Haller and Fontboté, 2009; Rusk et al., 2010; Zhao and Zhou, IOCG deposits worldwide, more than one model has been
2011). In situ sulfur isotope analyses using secondary ion mass proposed for the formation of Andean IOCG deposits based
spectrometry (SIMS) has been used to measure changes in the on the interpretation of data from several different systems.
isotopic signature (δ34S) within single pyrite grains in IOCG Models have invoked different fluid sources: (1) oxidized
deposits (Li et al., 2018), with the results revealing isotopic saline brines derived from evaporites that formed in conti-
variations at the grain scale. Comparisons between sulfur iso- nental back-arc basins during the Upper Jurassic and Lower
tope signatures and pyrite chemistry from IOCG deposits has Cretaceous (Barton and Johnson, 1996; Hitzman, 2000) and
produced mixed results varying from no clear correlation with (2) magmatic-hydrothermal fluids that evolved from tempo-
inferred mineralizing processes (e.g., Ernest Henry; Rusk et rally and spatially associated igneous intrusions (Marschik and
al., 2010) to potential evidence for least two different fluid Fontboté, 2001b; Sillitoe, 2003; Pollard, 2006; del Real et al.,
sources (e.g., Mina Justa; Li et al., 2018). 2018). Both transtensional (Arévalo et al., 2006; Groves et al.,
In this contribution we present a new and comprehen- 2010; Lopez et al., 2014; Richards et al., 2017) and transpres-
sive data set on pyrite chemistry from the Candelaria-Punta sional (Chen et al., 2013; del Real et al., 2018) tectonic set-
del Cobre district using three different microanalytical tings have been proposed coincident with the formation of
approaches: electron probe microanalysis (EPMA), laser abla- IOCG deposits in the Andean belt.
tion-inductively coupled plasma-mass spectroscopy (LA-ICP- The Candelaria-Punta del Cobre district includes 10 differ-
MS), and synchrotron micro-X-ray fluorescence (synchrotron ent active mines: Candelaria, Candelaria Norte, Alcaparrosa,
µ-XRF). EPMA and LA-ICP-MS data were used to evaluate Santos, Atacama Kozan, Granate, Punta del Cobre, Mantos
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 3

Antucoya Buey muerto

0 100 km
SCALE
Mantos Blancos

Atacam
24° S
ILE
a Fault Sy CH
CEAN

stem

Julia A
IN
PACIFIC O

Montecristo T
G EN
TALTAL AR

Casualidad
26° S
Todos los Santos
CHANARAL Carmen
Las Animas
as Santo Domingo
Mantoverde Porvenir
Portena
En torno a bella ester
Buena Aventura
Cerro Negro
Cerro Iman
COPIAPO Candelaria-
Punta del Cobre

Carrizal Alto
28° S
Boqueron Chanar
Los Colorados
VALLENAR Productora
Dos Amigos-Tricolor
Algarrobo
SYMBOLS AND AGES
Los Cristales > 160 Ma
Iron-oxide-apatite
deposits 150 – 160 Ma
Iron-oxide (Cu-Au)
deposits 130 – 150 Ma
LA SERENA Romeral Manto deposit
120 – 130 Ma
30° S Porphyry
120 – 130 Ma
Carmen de Andacollo Town
100 – 110 Ma
Atacama fault
system < 100
Panulcillo

Llahuin

El Espino
SOUTH AMERICA
PACIFIC OCEAN

El Soldado

Colliguay
ARGENTINA
Santiago
N
EA

CHILE
OC
NTIC

Fig. 1. Location of IOCG, iron apatite, manto, and porphyry

LA

deposits formed during the Upper Jurassic-Lower Cretaceous in

AT

northern Chile (from del Real et al., 2018).

4 DEL REAL ET AL.

del Cobre, Carola, and Las Pintadas (Fig. 2A). The main min- del Cobre Formation, which is divided into four main units:
eralizing event occurred at ~115 Ma (Mathur et al., 2002; del the lower andesite unit, the dacite unit, volcanic-sedimentary
Real et al., 2018) coeval with the emplacement of the Copiapó unit, and the upper andesite unit (Fig. 2A, B; Marschik and
batholith just west of the main deposits (Marschik and Söllner, Fontboté, 2001a; del Real et al., 2018). Several small mineral-
2006; Fig. 2A). Mineralization is mostly hosted in the Punta ized bodies are hosted in the overlying marine sedimentary

370000 375000 380000

(A)

6965000
Lar fa
1

2
ult

Me
len
dez
(B)
San G

Val

Copiapo Valley
ley
3

6960000
rego

4 NW SE
rio fa

Cerrillos Fm.
ult

Rocio

Cerrillos Fm.
e
on 6
fa

a rz
ult

he Candelaria
ss
co
jan
O 7
Pabellon Fm.

6955000
8
ult
fa
rte
i po Totoralillo Fm.

Chanarcillo Group
Pa
Ojan

or illa

Bandurrias Group
lin ar
cos f

ium
c am
ault

an rra
Tie

Nantoco Fm.
6950000

Abundancia Fm.
12
Upper andesite
Punta del Cobre Fm.

Volc.-sedimentary
Dacite
6945000

11 10

Lower
owLower
er A
Andesite
nde
andesite
site
9

Alluvial and coluvial Bandurrias Samples in the Chanarcillo Group

Punta del Cobre
deposit Group LP-1
Chanarcillo Group

Samples hosted above the manto horizon

Copiapo Batholith

Atacama gravels Abundancia/ Nantoco Los Lirios intrusive phase AD0093-14, ES0032-5, DH996-2
San Gregorio intrusive
Cerrillos Forma on Totoralillo/ Pabellon Samples from the manto horizon
phase
ES0032-15, AD0066-26
Deposit/prospect La Brea intrusive phase Samples hosted in lower andesite
Dacite dikes
DH996-21, DH996-23, LD1493-9
Fig. 2. (A) Simplified geologic map of the Candelaria-Punta del Cobre district with the main IOCG deposits (modified
from Arevalo, 1999). UTM coordinates are in datum PSAD56. Stars: 1 = Mantos de Cobre, 2 = Alcaparrosa, 3 = Santos, 4 =
Granate, 5 = Punta del Cobre, 6 = Carola, 7 = Candelaria, 8 = Atacama Kozan, 9 = Las Pintadas. (B) Stratigraphic column of
the geology in the Candelaria-Punta del Cobre district and the stratigraphic horizons sampled for this work. Stars represent
stratigraphic position of deposits used in this study: 10 = Candelaria, Alcaparrosa, 11 = Santos, 12 = Las Pintadas.
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 5

Abundancia Formation (e.g., Las Pintadas), which formed main economic mineralization. DH996-2 is from a thin
within the Chañarcillo basin. There is no evidence that the layer in the volcanic-sedimentary unit overlaying dacite
exposed parts of the Copiapó batholith generated ore fluids (Fig. 2B) ~100 m above the main orebody in the San-
(del Real et al., 2018), but minor vein-hosted mineralization is tos deposit. ES032-5 is from garnet-diopside-actinolite–
found in the batholith. altered upper andesite (Fig. 2B) ~200 m above the manto
The majority of the IOCG mineralization in the district is horizon just south of the Candelaria deposit. AD0093-14
hosted in the upper part of the lower andesite unit and the is from the dacite unit in the Alcaparrosa deposit, ~50 m
overlying volcanic-sedimentary and dacite units, all within the above the main mineralized orebody.
Punta del Cobre Formation. Mineralization is hosted in fault 3. Group 3: Samples LD1493-9, AD0357-3, AD0357-8,
zones, breccias, and specific lithologies. North-northwest AD0357-10, and AD009-14 are from late veins that cut the
faults are the dominant host for vertically extensive orebod- main manto mineralization. LD1493-9 is located within
ies. Fault-hosted orebodies are the dominant style of miner- the lower andesite (Fig. 2B) just below the contact with
alization in the eastern part of the district located along the the volcanic-sedimentary unit in the Candelaria deposit.
east side of the Copiapó Valley. Stratigraphically controlled Samples AD0357-8, AD0357-100 and AD009-14 are from
mineralization forms extensive stratabound orebodies (“man- the lower andesite below the main mineralized zone in
tos”), which are most abundant in the western part of the the Alcaparrosa deposit and sample AD0357-3 is from the
district where the most important lithological host, the vol- dacite unit above this mineralized zone. The latter four
cano-sedimentary sequence, is best developed (del Real et samples all contain anhydrite intergrown with pyrite or
al., 2018). Textural evidence suggests that the hydrothermal chalcopyrite.
system evolved and advanced upward over time (del Real et 4. Group 4: Sample LP-1 is from the main ore zone in the Las
al., 2018). The earliest event was dominated by magnetite- Pintadas deposit, which is located stratigraphically in the
actinolite in stratigraphically controlled mantos and exten- Abundancia Formation (Fig. 2B), the highest stratigraphic
sive zones of disseminated magnetite-actinolite in the deeper level for significant mineralization in the district.
parts of the Candelaria system. Magnetite-actinolite was sub-
sequently overprinted by chalcopyrite-pyrite–dominant min- Pyrite grains and aggregates from each sample were exam-
eralization in veins, fractures, and disseminated replacement ined using polarized-light microscopy at Cornell Univer-
with associated magnetite-actinolite-biotite-K-feldspar altera- sity supported by backscattered electron (BSE) imagining
tion (Marschik and Fontboté, 2001b; del Real et al., 2018). In through the EPMA at Syracuse University (EPMA analytical
addition to magnetite, iron oxides include widespread second- conditions explained in the next section). Exact locations and
ary magnetite (in the form of mushketovite) and hematite in more detailed descriptions of the alteration and mineraliza-
the upper part of some deposits. tion for each sample are reported in Table 1. Selected samples
were cut and mounted in 25-mm-diameter epoxy mounts and
Sampling and Analytical Methods polished down to 60-nm roughness using colloidal silica at the
Thirteen pyrite-bearing representative samples from the rock preparation laboratory at Syracuse University. The same
Candelaria-Punta del Cobre district were selected for this epoxy mounts were used for most of the analyses performed
study. The nine samples used for element mapping and in situ for this study (synchrotron µ-XRF, EPMA, LA-ICP-MS, and
sulfur isotope analyses were collected from Santos (samples SIMS).
DH996-2, DH996-21, and DH996-23), Alcaparrosa (samples Synchrotron µ-XRF
AD0093-14, AD006-26), Candelaria (samples LD1493-9,
ES032-5, and ES032-15), and Las Pintadas (sample LP-1). Synchrotron µ-XRF mapping was performed using beamline
The four remaining samples, which contain pyrite or chal- F3 at the Cornell High Energy Synchrotron Source (CHESS).
copyrite in textural equilibrium with anhydrite (samples Station F3, fed by a bending magnet and a double-multilayer
AD357-3, AD357-8, AD357-20, and AD009-14), are from monochromator, provided 14.5-keV incident X-ray energy for
the Alcaparrosa deposit. These were selected for whole-grain these scans. A four-element silicon detector (Vortex ME4)
sulfur isotope analysis coupled with whole-grain pyrite chem- with a Quantum Xpress3 digital signal was employed to col-
istry. The full sample suite represents different ore stages and lect the XRF signal. Under typical scan conditions of 20-µm
stratigraphic horizons: steps and 500-ms dwell time per pixel, typical signals reached
>250 kcps per channel with dead time <10%. A well-charac-
1. Group 1: Samples DH996-21, DH996-23, ES032-15, and terized, natural pyrite sample from the Los Colorados IOA
AD006-26 are from the main stage of mineralization, the deposit in northern Chile was used as an independent refer-
first two from structurally controlled orebodies and the ence for quantifying synchrotron µ-XRF measurements (del
other two from manto-style replacement mineralization. Real et al., 2019), since there is currently no commercial pyrite
Samples from structurally controlled mineralization are standard available. The scale factor (“monitor efficiency”) was
located within the lower andesite (Fig. 2B) near its contact adjusted to yield concentrations that matched the reference.
with the dacite unit (~<50–80 m) in the Santos deposit. The same scale factor was then applied to the subsequently
Samples from manto mineralization are from the volca- collected pyrite data sets in order to quantify concentrations.
nic-sedimentary unit (Fig. 2B) in the Candelaria deposit, Final quantification and correction of the data required appli-
stratigraphically above the lower andesite. cation of a variety of statistical methods as previously reported
2. Group 2: Samples DH996-2, ES032-5, and AD0093-14 in del Real et al. (2019). XRF maps were processed using the
were collected from mineralized areas peripheral to the open-source Praxes software package developed at CHESS
 6
Table 1. Pyrite Sample Descriptions, Including Paragenesis, Host Rock, and Location of Each Sample Used for this Study

Location collar

(Datum PSAD56)

Sulfur Depth in Z
Sample isotope drill hole surface
Trend Plunge
Sample Lithology classification analysis Pyrite type Description (m) North East (m) (°) (°)

ES032-15 Volcanic- Group 1, hosted In situ Single Layers of chalcopyrite-pyrite-pyrrhotite-magnetite- 899.8 6958496 373144 707 64 69
  sedimentary   in manto   grains   actinolite-feldspar intercalated with layers of
  unit   stratigraphic   biotite-feldspar; pyrrhotite is the main sulfide in
  horizon   the sample; sample paragenetically corresponds
  to the manto orebody in the Candelaria deposit
AD006-26 Volcanic- Group 1, hosted In situ Single Pervasive actinolite-magnetite alteration with patches 343.55 6962264 374337 484 90 112
  sedimentary   in manto   grains   of epidote and chlorite; minor disseminated
 unit  stratigraphic  feldspar; disseminated chalopyrite-pyrrhotite-
  horizon   pyrite; sample paragenetically corresponds to the
  manto horizon in the Alcaparrosa deposit
DH996-21 Lower Group 1, hosted In situ Aggregate Vein of magnetite-pyrite-chalcopyrite with feldspar 254.6 696117 376306 503 245.4 106
  andesite   in structurally   on the rim and chlorite-sericite patches in the
  controlled   center; chalcopyrite occurs in between pyrite
  orebody   grains or inside them as inclusions; surrounding
  host rock is obliterated to fine-grained biotite and
  magnetite; sample corresponds to the main
  mineralization orebody in the Santos deposit,
  hosted under the Dacite dome
DEL REAL ET AL.

DH996-23 Lower Group 1, hosted In situ Single Sample obliterated to fine-grained magnetite- 305.6 6961117 376306 503 245.4 106
  andesite   in structurally   grains   biotite (altered to chlorite)-feldspar-actinolite
  controlled   with patches of epidote; chalcopyrite and pyrite
  orebody   disseminated throughout the sample; sample
  corresponds to the main mineraization orebody
  in the Santos deposit, hosted under the Dacite dome
DH996-2 Volcanic- Group 2 In situ Single Sample altered to fine-grained feldspar-biotite 13.5 6961117 376306 503 245.4 106
  sedimentary   grains   (replaced to chlorite and sericite) with minor
  unit   disseminated chlorite and magnetite and epidote
  patches; chalcopyrite, pyrite, and magnetite
  disseminated throughout the sample; sample
  corresponds to shallow minor mineralization hosted
  on top of the Dacite dome in the Santos deposit
ES032-5 Upper Group 2 In situ Aggregate Upper andesite intercalated sediments obliterated to 455 6958496 373144 707 64 69
  andesite   garnet, diopside, and scapolite; patches of calcite
  and actinolite-plagioclase; plagioclase is altered to
  sericite; between the garnets there is pyrite with
  minor magnetite and chalcopyrite; sample corresponds
  to minor mineralization hosted on top of the
  Candelaria deposit


Table 1. (Cont.)

Location collar

(Datum PSAD56)

Sulfur Depth in Z
Sample isotope drill hole surface
Trend Plunge
Sample Lithology classification analysis Pyrite type Description (m) North East (m) (°) (°)

AD0093-14 Dacite dome Group 2 In situ Single Disseminated magnetite-feldspar with minor patches 185.7 6962294 374337 484 90 115
 grains   of actinolite and biotite and calcite; patches of
  magnetite-chalcopyrite-pyrite-biotite; sample
  corresponds to minor disseminated mineralization
  within the Alcaparrosa deposit
LD1493-9 Lower Group 3 In situ Single Chalcopyrite-pyrite-actinolite with minor feldspar 138.3 6956239 373365 544 245 99
  andesite   grains   and magnetite vein; surrounding host rock is
  obliterated to actinolite and magnetite with minor
  epidote and chlorite disseminated; sample corresponds
  to a late vein that cuts through the main mineralization
  and alteration in the Candelaria deposit
AD0357-10 Lower Group 3 Bulk NA Anhydrite-pyrite-chalcopyrite vein and minor chlorite 97.25 6962301 373935 471 90 1.5
  andesite   in the edges; sample corresponds to late vein that
  cuts the main alteration in the Lower Andesite
AD0357-8 Lower Group 3 Bulk NA Anhydrite-chalcopyrite vein; sample corresponds to 79.65 6962301 373935 471 90 1.5
  andesite   late vein that cuts the main alteration in the Lower
  Andesite, which is autobrecciated
AD0357-3 Dacite dome Group 3 Bulk NA Anhydrite-pyrite-chalcopyrite and minor chlorite- 25.65 6962301 373935 471 90 1.5
  epidote in the edges; sample corresponds to late
  vein that cuts the main alteration in dacite dome
AD009-14 Lower Group 3 Bulk NA Anhydrite-pyrite vein; sample corresponds to late 190.15 6961426 373935 493 0 0
CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE

  andesite   vein that cuts the main alteration in the Lower

  Andesite
LP-1 Abundancia Group 4 In situ Single Sample is heavily altered to biotite (partially altered 1,004 6947654 367142 1,004       Outcrop
  Formation   grains   to chlorite), fine-grained feldspar, minor epidote,
  and actinolite; disseminated chalcopyrite with minor
  pyrite and magnetite
7
8 DEL REAL ET AL.

(Dale, 2015). Praxes employs PyMCA libraries, developed at glass standards (GSC-1G, GSD-1G, and GSE-1G), and exter-
the European Synchrotron Research Facilities (ESRF) and nal standards (BHVO-1, MASS-1, and NIST612) were used
is widely used for XRF data processing, spectra fitting, and to monitor instrument drift and correct for elemental bias and
quantitative analysis (Solé et al., 2007). XRF is a full-spectral laser yield. Raw data were plotted against the element calibra-
technique, meaning that the signal is collected simultane- tion curves created using USGS glass standards to quantify
ously for all elements that fluoresce under the experimental the ablated areas. Each spot measurement was monitored as
conditions. The synchrotron µ-XRF has detection limits near it was acquired through a live cam in order to take note of any
the parts per billion level for most elements, while having no inclusions that were ablated during the analysis. Data were
problems in analyzing elements with major element concen- collected in time-resolved graphics mode to monitor possible
trations (e.g., Fe in pyrite). Detection limits calculated using compositional heterogeneities that might be present in the
the in-house reference were <1 ppm, with the exception of sample at the scale of the laser sampling and to monitor the
Co (~6 ppm). interelemental fractionation that can occur during laser abla-
tion analysis. The software package PlasmaLab was used for
EPMA selecting and monitoring the data integration space of each
Point data was acquired on transects across pyrite grains in point and element analyzed.
order to assess element variation in heterogeneous and zoned
crystals. Samples were carbon coated before being analyzed In situ sulfur isotope analysis
to avoid charge buildup. Transects of 10 to 15 analytical points In situ sulfur isotope measurements of pyrite and coeval
across grains were completed for each sample. Major and chalcopyrite reported in this study were conducted using the
minor element compositional analyses were performed at Syr- WiscSIMS CAMECA IMS 1280 large radius multicollector
acuse University using a Cameca SXfive field-emission elec- ion microprobe in the Department of Geoscience of the Uni-
tron microprobe with a Lab6 electron gun and five wavelength versity of Wisconsin, Madison. Analytical procedures were
dispersive spectrometers. For quantitative measurements, similar to those previously reported for sulfur two-isotope
the five wavelength dispersive spectrometers were tuned, (32S and 34S) analyses (Kozdon et al., 2010). Sulfide mounts
and elements were standardized using S and Fe on marcasite, in epoxy resin were Au coated (~30 nm). The secondary 32S–,
Sb on antimony, Co on cobalt, Ni on nickel silicide, Cu on 32S1H–, and 34S– ions were simultaneously collected by three

copper, Zn on sphalerite, As on gallium arsenide, Se on sele- Faraday cup detectors. In the routine analytical condition, a
nium, V on vanadium, and Ag on silver. The beam current was primary 133Cs+ beam with an intensity of ~1.6 nA was focused
adjusted to ~12,000 counts per second for analyte and X-rays to a 10-µm diameter with a Gaussian density distribution.
on gas-flow proportional counters. All imaging and quantita- The dish-shaped SIMS analysis pits formed by the Gaussian
tive measurements were performed using 15-kV accelerating focused beam have a depth of 1 to 2 µm. The standard
voltage. Measurements of unknowns were performed using UWPy-1 (Ushikubo et al., 2014) was used as a bracketing
a 20-nA beam current and a 2-µm beam diameter. Elements standard to monitor instrument stability and analytical spot-
were acquired using the following analyzing crystals: LIF for to-spot reproducibility. Grains of UWPy-1 were cast in the
Fe Kα, Co Kα, Ni Kα, As Kα, Se Kα, Cu Kα, and Zn Kα. center of each sample mount and were measured in at least
Counting time was 100 s for Ni Kα, As Lα, Se Lα, Co Kα, Cu four spots before and after every 10 to 12 analyses. Measured
Kα, and Zn Kα, and 20 s (10 s in two spectrometers) for Fe ratios of 34S/32S were reported in delta notation (δ34S) relative
Kα. Background times were determined by peak time divided to the Vienna Canyon Diablo troilite (VCDT). For δ34S, a cor-
by two. Because two backgrounds were measured (one on rection factor was determined for each of the UWPy-1 brack-
either side of the peak), the total background time measured ets by comparing the average measured value of the standard
equals the peak time. with its known value (16.04‰ VCDT). The precision of the
S-isotope analyses in this study, reported at the level of two
LA-ICP-MS standard deviation (2SD) varies between 0.05 and 0.4‰. For
Similar to EPMA, point data were acquired on transects chalcopyrite, Trout Lake chalcopyrite (δ34S: 0.3‰; Crowe
across pyrite grains in order to assess heterogeneous and and Vaughan, 1996) was used for estimating instrumental
zoned grains. Transects were designed to follow those ana- mass bias as a relative bias to UWPy-1 (–4.3‰) in the session.
lyzed by EPMA. LA-ICP-MS analyses were carried out at
the Queen’s University Facility for Isotope Research (QFIR) Bulk IRMS sulfur isotope analysis
using a XSeries 2 ICP-MS coupled to a New Wave/ESI Bulk sulfur isotope analyses were conducted at the Depart-
Excimer 193-nm laser ablation system. The LA-ICP-MS cali- ment of Geological Sciences, Queen’s University, Kingston,
bration was initiated by analyzing a U.S. Geological Survey Ontario. The weight range for samples varied between 0.3
(USGS) glass standard (GSD) to optimize He and Ar flow and 1.05 mg for pyrite and chalcopyrite. Anhydrite samples
through the ablation cell and the plasma torch. Optimum weighed between 0.71 and 1.12 mg. The sample size was set
plasma conditions were ensured by monitoring uranium to guarantee a minimum amount of sulfide or sulfate miner-
oxides (<0.6% UO/U). Point data were obtained using a beam als consistent with the purity of the sample. The sulfur iso-
diameter and spot measurement of 75 µm at a repetition tope composition for sulfides and anhydrite was measured
rate of 10 Hz, with a gas blank of 10 to 20 s. The laser beam using a MAT 253 isotope ratio mass spectrometer coupled to
was focused onto the surface of the sample, and the ICP-MS a Costech ECS 4010 element analyzer. The δ34S values were
determined the trace element concentrations in the ablated calculated by normalizing the 34S/32S ratios relative to VCDT
material. Analyses were bracketed by calibrations using USGS international standard, expressed in delta (δ34S) notation.
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 9

Total uncertainties are estimated to be better than ±0.2‰ for and minor pyrrhotite, which is only present in samples from
δ34S. Bulk trace element analyses of the same pyrite grains the manto horizon. Pyrite grains from group 2 (samples
measured by IRMS were performed using ICP-MS. The DH996-2, ES032-5, and AD0093-14; Fig. 3A, E) occur with
ICP-MS analyses were conducted at Queen’s University using more heterogeneous mineral assemblages. Samples DH996-2
an Finnigan MAT Element ICP-MS. The minerals analyzed and AD0093-14 exhibit patchy magnetite ± actinolite ± bio-
were handpicked from the same samples analyzed for bulk tite ± K-feldspar with minor chalcopyrite. Sample ES032-5,
sulfur isotope analyses. The weight range for samples varied closer to the contact with the Copiapó batholith, is charac-
between 30.2 and 121.7 mg. Samples were digested using terized by a pervasive garnet-diopside-actinolite-K-feldspar-
aqua regia, and reference materials CCu-1C and PTC-1 were scapolite alteration assemblage, with minor magnetite ±
measured along with the samples to ensure quality assurance specularite ± chalcopyrite. Samples with late veins in group
and quality control. 3 include LD1493-9 (Fig. 3B, F), where pyrite occurs ±
magnetite ± actinolite ± K-feldspar, AD0357-3, AD0357-8,
Results AD0357-10, and AD009-14 with pyrite in anhydrite-chalco-
pyrite-pyrite-epidote veins. The latter group of samples are
Textural relationships and paragenesis of pyrite samples all from the Alcaparrosa deposit, but similar veins occur in
Pyrite from the Candelaria-Punta del Cobre district is mostly many of the deposits in the Punta del Cobre district. Finally,
either subhedral to euhedral, is partly skeletal, and forms rela- pyrite grains from sample LP-1 (group 4) are dominated by
tively large grains, varying in size from ~2 to ~0.5 mm across. magnetite ± biotite ± K-feldspar ± specularite with chalcopy-
BSE images of pyrite grains reveal a granular texture around rite (Fig. 3C, G).
their border, suggesting growth of younger pyrite crystals with
minor chalcopyrite around early euhedral grains (Fig. 3A, B, Chemical zonation of pyrite
D). Some pyrite grains contain minor inclusions of chalcopy- The distribution of Se, As, Ni, and Co within pyrite grains
rite and magnetite (Fig. 3A, C). A complete description of the from the Candelaria-Punta del Cobre district was determined
mineral assemblages for each sample is presented in Table 1. by collecting synchrotron µ-XRF data for individual grains
Pyrite grains from group 1 (samples DH996-23, DH996-21, and generating elemental maps (Figs. 4–7). These elemental
ES032-15, and AD0066-26; Fig. 3D, H) are part of a mag- maps show chemical zonation and heterogeneities at the grain
netite ± actinolite ± biotite ± K-feldspar pervasive alteration scale for different trace elements in pyrite (e.g., core vs. rim),
assemblage with complete texturally destructive replacement and allow comparisons among samples throughout the district
of primary mineralogy. This assemblage contains chalcopyrite and mineralization stages. The elements that display the most

A B C D
pyrite magnete
magnete

pyrite pyrite
pyrite
chalcopyrite
chalcopyrite
pyrrhote
magnete chalcopyrite
500 μ 500 μ 500 μ 1000 μ
E F G H
magnete chalcopyrite pyrrhote
chalcopyrite chalcopyrite
acnolite pyrite
pyrite chalcopyrite pyrite

acnolite-biote-magnete
pyrite magnete
5 mm 5 mm feldspar 5 mm 5 mm

Fig. 3. BSE (A-D) and hand sample photos (E-H) of representative pyrite-bearing samples. (A) Sample DH996-2, euhedral
pyrite with pyrite regrowth on the edge of the main grain and chalcopyrite inclusions. (B) Sample LD1493-9, euhedral pyrite
from a vein with pyrite and chalcopyrite regrowth on the edge of the main grain. (C) Sample LP-1, euhedral pyrite sur-
rounded by chalcopyrite and with minor magnetite inclusions. (D) Sample ES032-15, euhedral pyrite with pyrite regrowth on
the edge of the main grain. (E) Hand sample of sample DH996-2, pyrite and chalcopyrite are disseminated in the volcanic-
sedimentary unit of the Punta del Cobre Formation. (F) Hand sample LD1493-9, pyrite-chalcopyrite-actinolite-magnetite
vein cutting main-stage actinolite-magnetite-K-feldspar pervasive alteration in the lower andesite of the Punta del Cobre
Formation. (G) Hand sample of sample LP-1, chalcopyrite-pyrite mineralization in the Abundancia Formation (lower part of
the Chañarcillo Group). (H) Hand sample of sample ES032-15, pyrite-chalcopyrite-pyrrhotite mineralization with pervasive
magnetite-biotite-K-feldspar-actinolite alteration in the volcanic-sedimentary unit of the Punta del Cobre Formation.
 Fe Se As Ni Co
(A) magnete 10

pyrite

AD006-26
pyrrhote 0.5 mm
0.5 mm 0.5 mm 0.5 mm 0.5 mm

% ppm ppm ppm %

8 24 40 56 72 0.0 40 80 120 160 200 0.0 100 200 300 400 500 0.0 1000 2000 3000 4000 5000 0.0 0.4 0.8 1.2 1.6 2.0

(B) Fe Se As Ni Co
chalcopyrite
pyrite

pyrrhote magnete

ES032-15
0.5 mm 0.5 mm 0.5 mm 0.5 mm 0.5 mm
% ppm ppm % %
8 16 24 40 56 72 0.0 40 80 120 160 200 0.0 100 200 300 400 500 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.4 0.8 1.2 1.6 2.0

(C) Fe Se As Ni Co

magnete pyrite
DEL REAL ET AL.

DH996-21
0.5 mm 0.5 mm 0.5 mm 0.5 mm 0.5 mm

% ppm ppm % %
8 24 40 56 72 0.0 40 80 120 160 200 0.0 1000 2000 3000 4000 5000 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.4 0.8 1.2 1.6 2.0

(D) Fe Se As Ni Co

chalcopyrite

pyrite

DH996-23
magnete
1 mm 1 mm 1 mm 1 mm 1 mm

% ppm ppm ppm %

8 24 40 56 72 40 80 120 160 200 1000 2000 3000 4000 5000 0.0 500 1000 1500 2000 2500 0.0 0.4 0.8 1.2 1.6 2.0

Fig. 4. Synchrotron-XRF element maps for pyrite samples from group 1. (A) Sample AD0066-23, disseminated pyrite, pyrrhotite, and magnetite from the manto horizon
in Alcaparrosa. (B) Sample ES032-15, pyrite grain surrounded by disseminated chalcopyrite, pyrrhotite, and magnetite from the manto horizon in Candelaria. (C) Sample
DH996-23, pyrite grain with magnetite from structurally controlled orebody in the Santos deposit. (D) Sample DH996-21, aggregate of pyrite grains contained in a vein
with magnetite and minor chalcopyrite from structurally controlled orebody in the Santos deposit. Areas of the image that do not include pyrite have been covered in
order to focus attention to elemental variation in pyrite.
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 11

significant variations at the grain scale are Ni, As, Se, and Co, on trace and minor elements with significant and consistent
and are described below for each unit. concentrations in the pyrite samples (e.g., Co, Ni, Cu, Zn, Pb,
Pyrite grains in samples from group 1, specifically those As, Sb, Se, Ag, and Cd). The Au data were discarded as the
from mineralized bodies within the manto stratigraphic hori- LA-ICP-MS analyses were carried out after the SIMS mea-
zon in Candelaria and Alcaparrosa (ES032-15 and AD066-26, surements, which had been coated with Au as part of the ana-
respectively; Fig. 4A, B) display a distinct zonation of Ni and lytical procedure.
As, where contents are higher in the core, with concentra- Boxplot diagrams of the combined LA-ICP-MS and EPMA
tions up to ~1 wt % and 500 ppm, respectively. Selenium data set (Fig. 8) show significant variations in most elements
contents display little zonation, although the concentration with the exception of Zn and Cu, where all samples display a
varies irregularly in some parts of individual pyrite grains (up similar range of values. Samples from group 1 (Fig. 8) have a
to ~100 ppm). The concentration of Co is low (~<100 ppm) in similar range of Cu, Ni, As, Se, Ag, and Cd values, but sam-
all of the pyrite samples from the manto horizon. The remain- ple ES032-15 (from the manto horizon) has higher values of
ing pyrite grains from group 1 (from fault-hosted orebodies; Pb and Sb contents than other samples. Pyrite samples from
samples DH996-23, DH996-21, and LD1493-9; Fig. 4C, D), group 2 (Fig. 8) have Pb, Sb, and Cd values similar to those
are zoned. In these samples the Co content is inversely cor- of sample DH996-2, also showing more variation in the con-
related with Ni and Se; i.e., zones with high Co contents have centration of most elements, especially Co, Se, As, and Ag.
low Ni and Se contents, and other zones contain elevated Ni Sample ES032-5 has higher Ni and lower Co concentrations
and Se concentrations and low Co contents. The distribution than the other two samples from group 2. Group 3 sample
of As is more erratic, having a positive correlation with Ni in LD1493-9 has elevated Pb, Sb, and Ag values compared to
sample DH996-21 and the opposite in sample DH996-23. all other samples (Fig. 8). Group 4 sample LP-1 (Fig. 8) has
Pyrite samples from group 2 are zoned, but chemical zon- the highest concentrations of Co and Se in the full data set
ing varies among the samples (Fig. 5). Samples ES032-5 and but low Ni and Pb concentrations, and Cd is below detection.
AD0093-14 display a negative correlation between Ni and Bulk analysis of pyrite in late veins from group 3 returned
Co and a positive correlation between Ni and Se, and both similar Se contents, whereas Ni varies from ~500 (sample
samples contain less Se than samples from group 1. Sample AD009-14) to ~3,000 ppm (sample AD357-10). Cobalt varies
AD0093-14 has a grain with a Co-rich core and another grain from ~70 (sample AD357-8) to 400 ppm (sample AD00914),
with a Ni-rich core and Co-rich rim (Fig. 5C). Sample ES032-5 and As is consistently low (<16 ppm) (Table 3). Sample
contains pyrite aggregates that include grains with elevated AD357-3 has Zn, Se, and Ni contents similar to those of sam-
Ni and low Co contents and grains with the opposite. The As ple AD0093-14 but lower As and Co contents.
contents in samples ES032-5 and AD0093-14 are lower than On a comparative basis for the complete data set, pyrites
in samples from group 1 and display a weak negative corre- from group 1 manto samples are enriched in Sb while
lation with Se. Sample DH996-2 contains grains showing a pyrite from late vein samples are enriched in Pb and Ag and
positive correlation between As and Co and a strong negative depleted in Se. The pyrite sample from the base of the Cha-
correlation between Co and Se, which is not observed in the ñarcillo Group (LP-1, group 4) in Las Pintadas deposit dis-
other two samples. plays different characteristics from the rest of the sample set,
Pyrite from the group 3 sample, LD1493-9 (Fig. 6), displays with the highest Co and Se contents and the lowest Pb and
a zonation pattern with Ni and As concentrated in the cores Cd contents.
of grains and the Co content increasing toward the rim of the
pyrite grains and aggregates. Selenium contents in this sample Stable sulfur isotope composition of pyrite
are low (~<15 ppm). In situ δ34S values were determined on the same pyrite sam-
Pyrite grains in group 4 sample, LP-1 (Fig. 7), show strong ples that were analyzed by synchrotron-XRF, EPMA, and
elemental zonation. The core and outer rim of the largest LA-ICP-MS. The results show a range of values from ~–2 to
grain in this sample have high Co (more than 1%) and very ~10‰ (Fig. 9A; statistical summary in Table 2 and the whole
high Se contents (up to 200 ppm), reflecting a strong positive data set in App. Table A2). The majority of the measured val-
correlation between Co and Se. The As content is highest in ues range between ~–1 and 2‰. The δ34S values calculated
the outermost rim where little Ni was detected (~<50 ppm). from in situ SIMS measurements on chalcopyrite grains adja-
cent to some of the pyrite grains display a similar range of val-
Major, minor, and trace element geochemistry of pyrite ues (Fig. 9B). Four pyrite samples have homogeneous sulfur
A total of 185 points in pyrite were measured for 58 elements isotope compositions (Fig. 10A, B, E, I), while the rest dis-
(Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, Ge, As, Se, play clear microscale variations; e.g., the δ34S values of sample
Rb, Sr, Y, Zr, Nb, Mo, Ag, Cd, In, Sn, Sb, Te, Cs, Ba, rare DH996-2 vary by 4‰ over <1 mm (Fig. 10F).
earth elements, Hf, Ta, W, Pt, Au, Hg, Tl, Pb, Bi, Th, U). Pyrite samples from the manto horizon, group 1, display
A statistical summary is presented in Table 2, and the data values between –2 and 2‰ and are homogeneous, with the
for elements with significant concentrations are provided in majority of δ34S values between 1 and 1.5‰ (Fig. 10A-D).
Appendix Table A1. LA-ICP-MS and EPMA analytical val- Samples from group 2 have heterogeneous δ34S values (Fig.
ues obtained from the same pyrite grains vary between 2 and 10F-H) with different samples showing different degrees of
10%, which is well within the variation attributed to the het- variation: sample DH996-2 –2.5 to 2.2‰, sample ES032-5
erogeneous nature of the pyrite grains, especially consider- –0.95 to 2.75‰, and sample AD0093-14 3.5 to 7.3‰. Pyrite
ing that each analytical method analyzed a slightly different grains from group 3 samples display homogeneous δ34S values
spot within the pyrite grains. Evaluation of the data focused that range between 9.4 and 10.06‰, and sample LP-1 from
(A) Fe Se As Ni Co 12

magnete

ES032-5
garnet pyrite

1 mm 1 mm 1 mm 1 mm 1 mm
% ppm % % %
8 16 24 40 56 72 0.0 40 80 120 160 200 0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.05 0.1 0.15 0.2 0.25 0.0 0.4 0.8 1.2 1.6 2.0

(B) Fe Se As Ni Co

pyrite

chalcopyrite

DH996-2
DEL REAL ET AL.

magnete
0.5 mm 0.5 mm 0.5 mm 0.5 mm 0.5 mm

% ppm ppm ppm %

0 24 40 56 72 0 40 80 120 160 200 0 1000 2000 3000 4000 5000 0 500 1000 1500 2000 2500 0 0.4 0.8 1.2 1.6 2.0

(C) Fe Se As Ni Co

chalcopyrite

pyrite

magnete

0.5 mm 0.5 mm 0.5 mm 0.5 mm 0.5 mm

AD0093-14
% ppm ppm ppm %
8 24 40 56 72 0.0 40 80 120 160 200 0.0 1000 2000 3000 4000 5000 0.0 500 1000 1500 2000 2500 0.0 0.4 0.8 1.2 1.6 2.0

Fig. 5. Synchrotron-XRF element maps for pyrite samples taken from group 2. (A) Sample ES032-5, aggregate of pyrite grains with garnet and magnetite from upper
andesite unit above de Candelaria deposit. (B) Sample DH996-2, pyrite grain with magnetite and chalcopyrite from above the dacite dome unit in the Santos deposit. (C)
Sample AD0093-14, pyrite grains with magnetite and minor chalcopyrite from dacite dome unit in the Alcaparrosa deposit. Areas of the image that do not include pyrite
have been masked in order to focus attention to elemental variation in pyrite.


Fe magnete Se As Ni Co

pyrite

chalcopyrite

LD1493-9
1 mm 1 mm 1 mm 1 mm 1 mm

% ppm % % %
0.0 16 32 48 72 0.0 40 80 120 160 200 0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.05 0.1 0.15 0.2 0.25 0.0 0.4 0.8 1.2 1.6 2.0
Fig. 6. Synchrotron-XRF element maps for pyrite sample from group 3. (A) Sample ES032-5, aggregate of pyrite grains with garnet and magnetite. (B) Sample DH996-2,
pyrite grain with magnetite and chalcopyrite (C) Sample AD0093-14, pyrite grains with magnetite and minor chalcopyrite. Areas of the image that do not include pyrite
have been masked in order to focus attention to elemental variation in pyrite.

Fe Se As Ni Co
magnete

pyrite
chalcopyrite
CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE

1 mm 1 mm 1 mm 1 mm 1 mm
% ppm ppm % %
0 24 40 56 72 0 40 80 120 160 200 0 100 200 300 400 500 0 0.5 1.0 1.5 2.0 2.5 0 1.0 2.0 3.0 4.0 5.0
Fig. 7. Synchrotron-XRF element maps for pyrite sample from group 4. Sample LP-1 was taken from the base of the Chañarcillo Group (Abundancia Fm) from Las
Pintadas deposit. A large pyrite grain is surrounded by chalcopyrite and minor magnetite. Areas of the image that do not include pyrite have been masked in order to
focus attention to elemental variation in pyrite.
13
14 DEL REAL ET AL.

Table 2. Statistical Average Values for Δ34S (‰) and Trace Elements (ppm) for Pyrite Samples

Sample n δ34S 2SD Co Ni Cu Zn Pb As Sb Se Ag Cd

ES032-15 38 1.33 0.28 1,191.36 5,045.09 83.48 7.18 38.46 120.38 22.35 71.70 0.59 0.11
AD006-26   9 1.52 0.12 2,664.30 2,458.43 475.54 29.85 2.37 355.49 0.06 44.55 2.30 0.06
DH996-21 28 –0.14 0.07 6,923.98 1,851.16 1,434.02 7.62 18.54 2,361.96 0.46 73.06 3.88 0.05
DH996-23 16 0.03 0.31 15,090.33 2,400.17 915.40 17.20 15.25 7,127.38 0.63 73.05 33.93 0.18
LD1493-9 14 9.60 0.14 4,707.15 2,377.69 6,444.97 14.83 821.36 3,442.11 1.48 16.85 24.85 0.03
DH996-2 22 –0.94 0.18 3,239.36 27.06 127.08 6.27 10.60 953.99 0.19 27.17 2.42 0.11
ES032-5 14 –0.15 0.40 504.38 1,440.12 829.36 55.71 302.65 209.87 0.28 42.08 9.71 0.72
AD0093-14 17 4.20 0.06 12,368.52 119.29 51.36 13.85 46.25 134.05 0.28 36.83 1.28 0.07
LP-1 27 1.44 0.14 18,348.89 56.86 31.10 3.35 0.41 236.63 0.03 100.21 0.93 -

Note: As some of the samples are zoned, some values within these grains can deviate from the numbers presented here; dash in otherwise-empty cell indicates
no data available

Table 3. Δ34S Isotope Values (‰) and Element Concentrations (ppm) for Whole Pyrite, Chalcopyrite, and Anhydrite Grains in Equilibrium

Sample Mineral Lithology δ34S Co Ni Cu Zn As Se

AD0357-10 Pyrite Lower andesite  7.7 232.54 3,161.83 1,770.12 155.24  5.55 11.95
AD0357-10 Anhydrite 18.1 - - - - - -
AD0357-8 Pyrite Lower andesite -  71.44 1,144.20 4,077.53  52.42  7.78 20.62
AD0357-8 Chalcopyrite   6.5 - - - - - -
AD0357-8 Anhydrite 17.7 - - - - - -
AD0357-3 Pyrite Dacite dome   4.0 121.58   722.74 4,057.00  40.03 16.92 20.84
AD0357-3 Anhydrite 16.0 - - - - - -
AD009-14 Pyrite Lower andesite 12.5 436.53   535.07   996.95  38.96  3.96 14.76
AD009-14 Anhydrite 21.4 - - - - - -

Note: dashes in otherwise-empty cells indicate no data available

group 4 is also homogeneous but with different δ34S values, 2017). These variations can be attributed to (1) changes in
1.17 to 2.1‰ (Fig. 10I). temperature of the hydrothermal fluid and the partitioning
The majority of the heterogeneous samples show an iso- of trace elements between coexisting mineral phases at dif-
topic zonation with increasing δ34S values from core to rim. ferent temperatures (Abraitis et al., 2004; Keith et al., 2016),
Samples that are composed of aggregates of small pyrite (2) changes in the redox and H+ activity conditions with the
grains (samples DH996-21 and ES032-5) display isotopic val- solubility of trace elements decreasing or increasing at dif-
ues that fluctuate, increasing and decreasing within a single ferent fO2 and pH, respectively (Thomson et al., 1993; Keith
transect. Chalcopyrite δ34S values (depicted in blue in Fig. et al., 2017), and/or (3) changes in the composition of the
10) correlate with the isotopic signature of the pyrite grains in hydrothermal fluid that will be reflected in the pyrite chem-
the same sample (Fig. 10B, D, E, G, H). Bulk δ34S analyses istry (Huston et al., 1995a; Abraitis et al., 2004; Tardani et al.,
obtained by Marschik and Fontboté (2001b) on chalcopyrite 2017).
from the district overlap with those obtained for this study Hydrothermal alteration mineral assemblages within the
(depicted in gray, Fig. 9B). Bulk δ34S analyses were carried Candelaria-Punta del Cobre district are relatively consis-
out on pyrite, chalcopyrite, and associated anhydrite from late tent both vertically and horizontally. The main alteration
veins, all of which appeared to be in textural equilibrium. The event from depth to the manto horizon consists of iron-rich
results provide sulfide δ34S values for pyrite and chalcopyrite calc-potassic mineral assemblages dominated by magnetite-
of 4 to 12.5‰ and sulfate δ34S values for anhydrite of 16 to biotite-K-feldspar-actinolite (del Real et al., 2018). At higher
21.4‰. (Table 3). The higher δ34S values in this group overlap stratigraphic horizons, alteration changes from magnetite to
with those from pyrite in the late vein sample LD1493-9. specularite dominant, and sodic-calcic alteration minerals
such as albite-garnet-diopside-scapolite-amphibole are con-
Discussion centrated in rocks of the upper andesite unit and units at the
base of the Chañarcillo Group, especially close to the Copi-
Evaluation of trace element signatures apó batholith (Marschik and Fontboté, 2001b; del Real et al.,
Despite its simple formula, pyrite can effectively incorporate 2018). The relative lack of distinct variations in the alteration
numerous trace metals in its structure, both in solid solution assemblages and paragenesis suggests that pH changes in the
and as nanoparticles (Reich et al., 2005, 2013; Deditius et al., hydrothermal fluids generally, or during fluid-wall-rock reac-
2009, 2011). Characterizing trace element variations within tions, were not spatially or temporally significant. Alteration
pyrite can serve as a monitor for changes in the hydrothermal mineralogy suggests a near-neutral pH for the hydrothermal
fluid evolution (e.g., Huston et al., 1995b; Large et al., 2009; fluids (Marschik and Fontboté, 2001b). Given this conclu-
Reich et al., 2013, 2016; Gregory et al., 2015; Tardani et al., sion, the following discussion focuses on potential changes in
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 15

temperature, fluid composition, and redox conditions as fac-

tors that may have affected pyrite chemistry and sulfur iso-
tope variation.
pp
m Significant variation of Se, As, Ni, and Co is observed
Cd
among samples and within individual pyrite grains from the
Candelaria-Punta del Cobre district (Figs. 4–7). Variations

Fig. 8. Box and whisker plot showing trace element concentrations determined by LA-ICP-MS integrated with EPMA. Cd was under detection limit in sample LP-1.
The central box represents 50% of data from quartile 1 (Q1) to quartile 3 (Q3); outlier circles and triangles indicate the data that is farther than 1.5 (Q3-Q1) from the box.
of other elements such as Pb and Sb are present in specific
samples (Fig. 8), and we attribute these largely to small-scale
inclusions (Abraitis et al., 2004). Although there is no signifi-
pm cant variation of the alteration paragenesis in the district, the
chemical zonation observed in pyrite from most samples sug-
p
Ag

gests that the changes in the conditions of the hydrothermal

fluid involved multiple events, at least at a local scale. Although
mineral grain kinetic effects can influence mineral growth and
chemical zoning (Jamtveit, 1991; Putnis et al., 1992), these are
not considered important for variations in pyrite chemistry in
m
pp
Se

this study because no quasi-cyclic zoning (or oscillatory zon-

ing) in the chemical composition was observed in any of the
samples used in this study (Figs. 4–7).
Selenium is able to substitute stoichiometrically for S in the
pyrite structure. Several researchers have noted increasing Se
m
pp
Sb
concentration in pyrite with an increase in temperature (Hus-
ton et al., 1995b; Revan et al., 2014; Krumm et al., 2015; Keith
et al., 2017) or a rise in fO2 (Huston et al., 1995b; Large et
al., 2014). Selenium varies significantly in the pyrite samples,
but in most cases, it has a general positive correlation with
pm
As
p
Ni (Fig. 11A). Since incorporation of Ni into pyrite tends to
be more efficient at higher-temperature conditions, the posi-
tive Ni-Se correlation suggests that the incorporation of Se in
pyrite may also be at least partly controlled by temperature
(Lehner et al., 2006).
pp
m
Pb
Arsenic may substitute for S nonstoichiometrically in the
pyrite structure—a process requiring the substitution of ions
of differing net charge polarity. The incorporation of As into
the pyrite lattice may lead to distortion of the pyrite structure,
increasing defect formation (Abraitis et al., 2004). Structural
distortion may allow other trace elements to enter the pyrite
m
pp structure, resulting in a positive correlation between As and
other trace elements (e.g., Griffin et al., 1991). Arsenic con-
Zn

centration in pyrites analyzed in this study does not correlate

with Ni or Se, but in some of the pyrite grains there is a posi-
LP-1
Group 4

The whiskers include the extreme outlier values.

tive correlation between Ni and Co (e.g., samples DH996-2

and DH996-23; Figs. 5, 6, 11B). Samples AD0093-14 and
LP-1, however, are distinctly different, with no As-Co corre-
LD1493-9

m
pp
Cu
lation and highly elevated Co contents. Based on the avail-
Group 3

able data in this study, the As content in pyrite does appear to

provide exclusive control on the concentration of other trace
elements.
AD0093-14
DH996-2

ES032-5

A positive correlation between As and Co is expected when

Group 2

m
pp the hydrothermal fluid contains both elements (Abraitis et al.,
2004), and therefore, the lack of correlation between Co and
Ni

As, as observed in samples AD0093-14 and LP-1, suggests a

structure hosted
DH996-23

DH996-21

decoupling of these elements. This decoupling would poten-

tially be associated with changes in the composition and/or
nature of the hydrothermal fluid, as has been proposed for Cu
and As in high-sulfidation systems (Deditius et al., 2009; Tar-
AD006-26
“manto” hosted
ES032-15

m
pp

dani et al., 2017). Sample LP-1 (group 4) from the main min-
Co
Group 1

eralization event in Las Pintadas deposit has well correlated

Co and Se contents compared to the rest of the sample set,
0.001
0.01
10000

0.1
1000

100

1
10

log (ppm) together with lower Ni concentrations (Table 2). Las Pintadas
16 DEL REAL ET AL.

A δ34 S histogram of pyrite samples deposit is ~5 km south of the main cluster of IOCG deposits
70
in the Punta del Cobre district (Fig. 2) and is hosted at the
66
base of the Chañarcillo Group, at a higher stratigraphic level
n=185 than the other major deposits. The differences in the pyrite
60
chemistry suggest that the hydrothermal fluid responsible
50 for mineralization in Las Pintadas may have had a distinct
composition compared to IOCG mineralization in the rest of
Frequency

39
40 the district. Sample AD0093-14 from group 3 is hosted the
dacite dome unit above the manto horizon in the Alcaparrosa
30 deposit, and therefore, it seems unlikely that pyrite grains in
22
this sample formed under identical conditions to those for
20 17
sample LP-1 at Las Pintadas. The decoupling of As and Co in
10 8 8 9 10 both samples, however, suggests that some similar processes
3
2 3 influenced incorporation of As and Co in pyrite in both areas.
1
0
-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 Co and Ni concentration and Co/Ni ratios
δ S (‰)
34 Both Co and Ni may substitute stoichiometrically for Fe in
pyrite, reflecting the fact that Co2+ and Ni2+ have a similar
δ S histogram of chalcopyrite samples
34
ionic radius to Fe2+ (Tossell et al., 1981; Abraitis et al., 2004).
B 16
16
n=20 The Co and Ni concentrations reported here for pyrite from
6 the Candelaria-Punta del Cobre district are considerably
n=23
higher than in other hydrothermal systems, including por-
Frequency

5 4
4 phyry copper deposits and epithermal Au-Ag systems (Reich
5 3 et al., 2013; Deditius et al., 2014; Franchini et al., 2015), with
2 2 2
1 2 1 1 samples from the manto horizon having Ni concentrations
1 locally >1 wt % (Fig. 8). The concentration of Co and Ni is
0
-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 higher in mafic magmas than in intermediate magmas (Taylor
δ S (‰)
34 et al., 1969; Nicholls et al., 1980; Rudnick and Taylor, 1987;
Longhi et al., 2010; Nadeau et al., 2010) and significantly
Fig. 9. δ34S histogram of in situ measurements obtained from pyrite (A) and
chalcopyrite (B). The highest concentration of values concentrate between
higher than felsic magmas (Gülaçar and Delaloye, 1976; Zhao
–1 and 2 δ34S. Samples depicted in gray correspond to analysis carried out by et al., 2011). Therefore, the andesitic and dacitic volcanic host
Marschik and Fontboté (2001b). rocks of the Punta del Cobre Formation are an unlikely source

ES032-15 DH996-21 DH996-23 LD1493-9

- 1.1
- 0.1 3
2.15
1.30 0.9 -0 0.67 3 9.93
-0.80 0.03 -0 .38 9.96
.3 9.98
-0.86 -0.13 6
-0.90 -0.04 10.02
-0.82 10.18 9.91
-0.80 0.32 -0.31 9.89
0.46 -0.54 -0.05 10.06
0.8 5

1.18
0.94
0.945
1.187
1.0 4

-0.29
1.30

1.586
1.49
1.104

0.21
1.4

1.1

0.57 9.94
0.

-0.12 0.47 8.61 9.78

0.49
-1.29 -0.45 9.45
1.5 8
1.75

1.262
1.72
1.24
1.47
1.39

1.541
1.28
1.68
1.47

1.4
1.6

-0.25 -1.00

9.41
9
0.9 0.94

1.31
1.18
9
1.13 0.

1.74 1.3

-0..509
-0.87
-0.91
0. 1

3
-0..331
-0 5

-0.967
-0.95
-1.91
-0.54

-0. 0
-0 4
-0.69
-0.83

2.981
-1.01
2.0

9.85
0.73
1.03

-0.58 10.03
1 mm 1 mm 1 mm 0.5 mm
-0.19

AD006-26 DH996-2 ES032-5 AD0093-14 LP-1

3.16
1.33 -1.98 0.32
1.31 1.40 -2.11 -1.79 -0.80
-1.95 -2.18 -0.82 7.01
1. 35

1.64 1.15 -0.65 3.99 7.36

1 29
1.1.18

-2.20 -2.20
1.

1.2 7

-0.95
2.4 23 1.3 1.510

1.57 4.05
1.26 1.49 3.40
-0.40 3.80
3.49
1.5 3 1.474

0.94
3.71
1.5 2

2.50 -0.82
1.6 6
-2. 6
-2.16
-2..50
1.939

4.078

6 2.1
1.5

-2 30

8
4

3.83 1
4

1. 5

2.32 -1.01 1 .25

3.7

1.2
3.
3

3.74
-0.73 3.74 3.66 1 .54
-0.75 3.86 3.82 3.94 1.4.52
-0.44 3.73 1. 3
1.68 0.10 3.47 4.35 3.78 1. 23
1.65 2.75 3.72 1.632
-1. 92
-0.10

0.84 1.86
-1.93
-1.44
0.384
2.

1.25 3.74 7
4.37
7

0.5 mm 0.85 1 mm 0.5 mm 0.5 mm 1 mm

-3 1 4 7 >10
δ34S
Fig. 10. δ34S
in situ measurements of pyrite and chalcopyrite grain samples for this study. Chalcopyrite values are depicted in
blue. Point colors are based on color scale. The same points were measured using LA-ICP-MS for pyrite chemistry.
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 17

(A)

10000

1000
Ni (ppm)

100

10

10 20 30 100 200

(B) Se (ppm)

10000
Co (ppm)

1000

100

10 20 30 100 200 1000 10000

As (ppm)
Group 1 Group 2 Group 3 Group 4 Pyrite type
Structure hosted Manto hosted LD1493-9 Rim
DH996-23 ES032-15 DH996-2 Core
AD357-3 LP-1
DH996-21 ES032-5 Grain aggregates
AD006-26 AD357-10
AD0093-14 Whole grain
AD357-8
AD009-14
Fig. 11. Element variation diagrams from LA-ICP-MS and EPMA analysis on pyrite grains showing a positive trend between
Ni and Se (A) and Co and As (B). In both diagrams data points from sample LP-1 from the main mineralization event from
Las Pintadas are off the positive trend formed between the other samples, and in the diagram of As and Co sample AD0093-
14 is also off trend together with LP-1

for the high Ni and Co concentrations observed in pyrite. The and volcanic settings (Bralia et al., 1979), and Co/Ni ratios
Ni-Co contents are most likely derived from deep mafic rocks between ~1 and 10 are characteristic of magmatic-hydrother-
via hydrothermal leaching or magmatic-hydrothermal fluids mal deposits (Bajwah et al., 1987; Reich et al., 2016).
released from mafic magmas. The relatively restricted range of Co/Ni ratios of pyrite from
Variations in Co/Ni ratios in pyrite have been used as a the Candelaria-Punta del Cobre district suggests incorpora-
proxy for classifying the origin and source of hydrothermal tion from a fluid of a similarly restricted ratio. Minor changes
mineral deposits, since both elements may be incorporated in the Co/Ni ratio from core to rim in pyrite from some sam-
equally, conserving the ratio in associated hydrothermal fluids ples (e.g., DH996-23) are not sufficient or uniform enough
(e.g., Bralia et al., 1979; Campbell and Ethier, 1984; Bajwah to suggest broad changes in fluid composition. The dominant
et al., 1987; Large et al., 2009; Koglin et al., 2010; Reich et al., range of Co/Ni ratios between ~1 and 10 (Fig. 12) is consis-
2016). Previous research using Co/Ni ratios determined that tent with pyrite of a magmatic-hydrothermal origin as defined
low Co/Ni ratios (<1) are characteristic of pyrite in mineral by Reich et al. (2016). The Co/Ni ratios higher than 100 occur
deposits that formed at or below the seafloor in sedimentary in pyrite grains with very low Ni concentrations in samples
18 DEL REAL ET AL.

0.1 0
10

Co %
i=
Co/N
0.01 i=10
Co/N i=1
0.001 C o/N 0.1
i=
Co/N
0.0001
0.0001 0.001 0.01 0.1 1
Ni %
Group 1 Group 2 Group 3 Group 4 Pyrite type
Structure hosted Manto hosted LD1493-9 LP-1 Rim
DH996-23 ES032-15 DH996-2
AD357-3 Core
DH996-21 AD006-26 ES032-5 Grain aggregates
AD357-10
AD0093-14 Whole grain
AD357-8
AD009-14

Fig. 12. Co versus Ni variation diagram and Co/Ni ratios of the pyrite measurements. The Co/Ni ratios of pyrite from the
Candelaria-Punta del Cobre district determined in this study range between ~0.1 and 100, where the majority of samples
range between 1 and 10, consistent with pyrite of a magmatic-hydrothermal origin. Samples with Co/Ni ~100 reflect pyrite
grains with very low Ni concentrations formed at shallower stratigraphic levels.

from higher stratigraphic levels (e.g., LP-1 and DH996-2). Sakai, 1984), (4) andesites typically have more positive δ34S
The composition of wall rocks or formation at lower tempera- values ~2.6‰ (Rye et al., 1984), and (5) the oceanic sulfur
tures, both potentially consistent with their stratigraphic posi- cycle that includes euxinic black shales tends to display nega-
tion, may have influenced the composition of pyrites in these tive values of δ34S <–10‰ (Chambers, 1982; Strauss, 1997).
samples. Although most of the δ34S values for pyrite samples from
The range of Co/Ni ratios from pyrite in the Candelaria- the Candelaria-Punta del Cobre district concentrate between
Punta del Cobre district is similar to the range found at the –2 and 2‰, the presence of samples with values as high as
Mantoverde and Ernest Henry IOCG deposits (Benavides et 12‰ (Fig. 9) requires variations in fluid conditions or the
al., 2007; Rusk et al., 2010) and at the Los Colorados IOA source of sulfur. The limited variations within the bulk of the
deposit (Reich et al., 2016). At Los Colorados, these ratios data may reflect minor variation in the source of S, interac-
have been interpreted as an indicator of the mafic affinity for tion with different host rocks, or slight temperature changes.
the magmatic-hydrothermal fluid source (Reich et al., 2016). A Changes in the fO2 of a hydrothermal fluid would not affect
mafic magmatic source for fluids is also suggested for the Can- the isotopic composition except when pyrite or other sulfide
delaria Punta del Cobre system, which is further supported by minerals are in equilibrium with Fe oxides, such as magnetite
Cl isotope data indicative of a mafic- or mantle-derived origin and/or hematite (Ohmoto, 1972; Rye and Ohmoto, 1974). In
for the hydrothermal fluids responsible for mineralization in these cases, δ34S values may differ from the original δ34S of the
the Candelaria deposit (Chiaradia et al., 2006). fluid source, and variations will reflect a variation in fO2 of the
ore-forming fluids. The Fe oxide alteration mineralogy associ-
Origin of δ34S variation ated with the Candelaria-Punta del Cobre deposits includes
The isotopic composition of sulfur in hydrothermal minerals the presence of magnetite, mushketovite (specular hematite
is strongly controlled by the temperature, fO2, and pH of the replaced by magnetite), and specular hematite, where specu-
hydrothermal fluids (Ohmoto, 1972; Rye and Ohmoto, 1974). larite characterizes the late events and the highest levels of
The isotopic ratios of sulfur (δ34S) in sulfide minerals have the hydrothermal system (Marschik and Fontboté, 2001b;
been successfully used to interpret the origin of ore deposits del Real et al., 2018). Changes in the Fe oxide mineralogy
(Seal, 2006, and references therein). The sulfur isotope sig- may reflect variation in fO2 and temperature conditions of the
natures of different ore-related geologic environments range hydrothermal fluid (Ohmoto, 2003; Otake et al., 2010).
from strongly positive to strongly negative with the following Sulfide minerals in equilibrium with pyrrhotite are more
breakdown: (1) marine evaporate sequences tend to display likely to possess δ34S values close to the initial δ34S values
values of δ34S >10‰ (Claypool et al., 1980; Strauss, 1999), (2) of the source fluid when the temperature of mineralization
mantle sulfides have δ34S between ~0 and 1.0‰ (Seal, 2006), is >200°C (Ohmoto, 1972). The mineral assemblage associ-
(3) continental and island-arc basalts are similar or slightly ated with group 1 samples from the manto horizon comprises
more positive than mantle sulfides at δ34S ~1.0‰ (Ueda and intergrowths of chalcopyrite-pyrite-pyrrhotite-magnetite, and
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 19

therefore it is likely that sulfide δ34S values reflect those of the more oxidizing conditions, above the Ni-NiO redox fO2 buf-
source hydrothermal fluids. Values for these samples range fer (Kress and Carmichael, 1991). Therefore, Ni/Se ratios
between ~1 and 1.5‰, consistent with a magmatic source of could potentially be used as a proxy indicative of oxidation
sulfur. Magmatic sulfur could be derived directly from a mag- states, with higher ratios suggesting more oxidized conditions.
matic-hydrothermal fluid or by leaching magmatic sulfur from Integrating these ratios with δ34S data (also redox sensitive)
the volcanic sequence in the footwall. Pyrite Co/Ni ratios of may provide a more robust assessment of redox conditions,
~1 for pyrite in samples from the manto suggest a mafic mag- where δ34S values will tend to be lower at higher fO2 values
matic source, but this is unlikely to be from the intermediate (Ohmoto, 1972). Nickel/selenium ratios plotted against δ34S
volcanic rocks of the Punta del Cobre Formation, as discussed (Fig. 13) show a rough trend where some of the samples with
above. A magmatic-hydrothermal origin for the sulfur and the higher Ni/Se ratios (e.g., LD1493-9, AD357-10, AD357-8)
Co/Ni ratios of the pyrite, and by inference the mineralizing have higher δ34S values, and some of the samples with lower
fluids, is therefore preferred. In contrast, pyrite from group 3 Ni/Se ratios (e.g., DH996-2 and LP-1) have lower δ34S val-
late vein samples LD1493-9 and AD0357-14 (anhydrite-bear- ues. Data from the remaining samples (ES032-5, DH996-21,
ing vein) show δ34S values >10‰, suggesting the participation DH996-23, AD006-26, and ES032-15) do not fall clearly on
of a late hydrothermal fluid with a more positive δ34S, poten- this trend but can be interpreted as a second potential trend.
tially from an evaporitic source, as proposed for mineraliza- If these trends are distinct, it is possible that the same pro-
tion in the Marcona/Mina Justa district (Li et al., 2018). cess is controlling the correlation between Ni/Se ratios and
Nickel and Se are redox- and temperature-sensitive ele- δ34S superimposed on fluids with different background Ni/Se
ments, and therefore, changes in Ni/Se ratios could be an values. Higher Ni/Se ratios and lower δ34S values are indica-
indicator of changes in temperature, mineral phases, or tive of a more oxidizing environment; therefore, this general
redox, with potentially more than one factor being important trend may reflect fO2 variation. The presence of mushketovite
(as discussed previously). The positive correlation of Ni and throughout the district may also relate to changes in the fO2
Se (Fig. 11) may reflect high-temperature conditions, but condition of the hydrothermal fluid (Ohmoto, 2003; Otake et
the high Se and low Ni contents observed in sample LP-1 al., 2010). We propose that the formation of mushketovite is
from group 4 suggest that temperature is not the exclusive related to redox changes, as inferred temperatures for sulfide
control on the concentration of these elements. Changes formation (explained in the next section) are too high for non-
in fO2 conditions potentially also played a role in control- redox transformations between Fe oxides (>250°C). Never-
ling the content of Ni and Se in pyrite. In reduced environ- theless, fO2 variation may not be the only factor affecting the
ments, Se occurs predominantly as Se–2 (when Se–2 is reduced range of δ34S values observed in the sample set, as changes in
from Se0; Johnson, 2004), whereas Ni occurs as Ni+2 under the temperature can also affect δ34S isotope values (Ohmoto,

AD357-10
ES032-15 LD1493-9
DH996-23
100
DH996-21 AD006-26
AD357-8

ES032-5 AD357-3 AD009-14

10
Ni/Se

1
AD0093-14
DH996-2
LP-1

0.1

-2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12

δ34S(‰ )
Group 1 Group 2 Group 3 Group 4 Pyrite type
Structure hosted Manto hosted LD1493-9 LP-1 Rim
DH996-23 ES032-15 DH996-2
AD357-3 Core
DH996-21 AD006-26 ES032-5 Grain aggregates
AD357-10
AD0093-14 Whole grain
AD357-8
AD009-14
Fig. 13. Pyrite chemistry integrated with δ34S measurements; higher Ni/Se ratios roughly correlate with a higher δ34S. See
text for further discussion.
20 DEL REAL ET AL.

1972). As mentioned above, both Ni and Se are temperature- Table 4. Sulfide-Sulfate and Sulfide-Sulfide Temperatures
sensitive elements, while Se has been proposed to be more Calculated Using Δ34S Values
sensitive to temperature than redox changes (Keith et al., Sample Geothermometer T (°C ) ± (°C) Host rock
2017); therefore, varying temperature in addition to fO2 may
also explain the poorly defined trends. AD0357-10 Pyrite-anhydrite 533 50 Lower andesite
Redox variations within a hydrothermal system have been AD0357-8 Chalcopyrite-anhydrite 534 50 Lower andesite
AD0357-3 Pyrite-anhydrite 480 50 Dacite dome
identified as a key process for Cu-Au transport, specifically AD009-14 Pyrite-anhydrite 606 50 Lower andesite
in porphyry copper systems (Sun et al., 2004, 2013). In mag- AD0093-14 Pyrite- chalcopyrite 394 50 Dacite dome
netite-rich systems, mineralization may be associated with DH996-23 Pyrite- chalcopyrite 552 50 Lower andesite
magnetite crystallization, accompanied by decreasing pH and ES032-15 Pyrite- chalcopyrite 450 50 Volcanic-
corresponding increase in fO2. Once sulfate reduction lowers  sedimentary
 unit
pH sufficiently and the fO2 reaches the hematite-magnetite
oxygen fugacity buffer, hematite forms and pH increases for a Notes: Sulfide-sulfate temperatures were calculated using equations by
given fO2. The oxidation of ferrous iron during the crystalliza- Ohmoto and Lasaga (1982); sulfide-sulfide temperatures were calculated
tion of magnetite and hematite would be the causal process using equations by Kajiwara and Krouse (1971)
for sulfate reduction and consequent mineralization (Liang et
al., 2009; Jenner et al., 2010; Sun et al., 2013).
In the case of the Candelaria-Punta del Cobre district, the stratigraphically deeper samples. Although limited, these
deep-seated high-temperature hydrothermal fluids could results are consistent with cooling of an ascending hydrother-
have channeled upward through structures at sufficient flux mal fluid.
rates and fluid/rock ratios to minimize interaction with the Our results correlate well with previous calculations that
host rock, as proposed for the Raúl Condestable deposit (De estimated that the core of the Fe mineralization in the district
Haller and Fontboté, 2009). These fluids may have been rela- formed between 500° and 600°C (Marschik and Fontboté,
tively reduced, hence precipitating the barren early magnetite 2001b) and that Cu-Au mineralization formed from 400° to
alteration observed in the district (del Real et al., 2018). As 500°C (Hopf, 1987). Limited fluid inclusion analyses from
temperature decreased, hydrothermal fluids would become quartz and anhydrite in the Candelaria deposit gave homoge-
more oxidized, reaching hematite stability, between 500° nization temperatures in the range of 330° to >470°C (Ulrich
and 250°C (Helgeson et al., 1978; Myers and Eugster, 1983; and Clark, 1999; Marschik and Chiaradia, 2000), which is
Giggenbach, 1997; Einaudi et al., 2003). As proposed for por- consistent with the range of temperatures calculated using
phyry deposits, the formation of hematite would increase the sulfide-sulfate and pyrite-chalcopyrite δ34S geothermometry
pH for a given fO2 causing sulfate reduction and sulfide precip- described above. Temperatures calculated for mineraliza-
itation (Liang et al., 2009; Jenner et al., 2010; Sun et al., 2013). tion in the Candelaria-Punta del Cobre district are similar to
The replacement of hematite by magnetite (forming mush- those calculated for the Mantoverde deposit in Chile and the
ketovite) implies that conditions became more reduced, pos- Marcona/Mina Justa district in Peru using oxygen isotope geo-
sibly as a result of the arrival of multiple pulses of deep-seated thermometry (Chen et al., 2011) and are significantly lower
hot hydrothermal fluids. Alternative processes, including the than estimates for IOA deposits and deeper segments of the
influence of local wall rocks or mixing with other fluids, have Candelaria deposit (e.g., 600°–>850°C in Los Colorados, El
been suggested as mechanisms for mushketovite formation Romeral, and Cerro Negro Norte in Chile; 400°–>900°C in
(De Haller and Fontboté, 2009), but neither clearly explains Kiruna, >700°C below the Candelaria deposit; Jonsson et al.,
the concetration of musketovite in some areas or its overall 2013; Knipping et al., 2015b; Bilenker et al., 2016; Rojas et
widespread distribution in all of the deposits in the district. al., 2018, Salazar et al., 2019; Rodriguez-Mustafá et al., in
press). Temperatures calculated for the manto horizon, where
Temperature and origin of hydrothermal fluids a significant part of the high-grade mineralization is concen-
The δ34S values of sulfide-sulfide pairs and sulfide-sulfate trated, partially overlap with the temperatures proposed for
pairs that crystallized in equilibrium in hydrothermal systems mineralization in porphyry copper systems (~300°–420°C;
can be used to estimate the temperature of the hydrother- Hedenquist and Lowenstern, 1994; Heinrich et al., 2008).
mal fluid at the time of mineralization (Kajiwara and Krouse, Ore deposition is interpreted to occur at ~400°C in porphyry
1971; Ohmoto and Lasaga, 1982). Table 4 lists the tempera- copper environments, with efficient cooling of the hydrother-
tures calculated using δ34S of aqueous sulfide-sulfate (pyrite mal fluid being the key to high Cu-Au grades (e.g., Heinrich
or chalcopyrite with anhydrite; Ohmoto and Lasaga, 1982) et al., 2008). A similar interpretation can be suggested for
and in situ sulfide-sulfide pairs (pyrite-chalcopyrite in equilib- IOCG deposits in the Candelaria-Punta del Cobre district, as
rium; Kajiwara and Krouse, 1971). The range of temperatures the highest grades are associated with temperatures similar to
calculated using δ34S results of sulfide-sulfate and sulfide-sul- those in porphyry Cu deposits.
fide pairs from the main mineralization event hosted in the The high δ34S values (>10‰) obtained for pyrite samples
lower andesite unit below the manto horizon is 530° to 600 ± from group 3, as previously mentioned, suggest the addition
50°C. The range of temperatures calculated using δ34S results of an external fluid. Ratios of Se/S can be used as a proxy
of sulfide-sulfate and sulfide-sulfide pairs from the dacite for tracing hydrothermal fluids in ore systems (e.g., Huston
dome unit and volcanic-sedimentary unit (both stratigraphi- et al., 1995a; Fitzpatrick, 2008; Li et al., 2018), with differ-
cally correlated with the manto horizon) varies between 394° ent fluid sources having distinct Se/S ratios. Seawater (or
and 480 ± 50°C, lower than the temperatures calculated from basinal water ultimately derived from evaporated seawater)
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 21

has an average δ34S value ~21‰ and a mass SSe/SS ratio values in the Candelaria-Punta del Cobre district in pyrite
of ~0.0500 to 0.25 × 10–6, whereas magmatic-hydrothermal from group 3 samples (LD1493-9 and AD009-14) suggest
fluids have a typical δ34S value of ~0 with a range of –2 to that basin-derived fluids circulated into the Candelaria sys-
2.4‰ for a fluid that exsolved from a crystallizing magma tem toward the end of the main mineralizing event. Further
of andesitic composition, and a mass SSe/SS ratio of ~120 work is needed to test this hypothesis. As all main-phase
to 500 × 10–6 (Huston et al., 1995a; Seal, 2006, and refer- mineralization samples fall close to, or within, the magmatic-
ences therein; Fitzpatrick, 2008). When plotting Se/S ratios hydrothermal range, it is reasonable to interpret the main
against δ34S isotope values (Fig. 14), the majority of the ore depositional event as being dominantly of magmatic-
pyrite samples from group 1 fall within the range of mag- hydrothermal origin.
matic-hydrothermal fluids (defined by Fitzpatrick, 2008) Most of the pyrite analyses in sample DH996-2 plot off the
although showing a spread to lower Se/S ratios. The samples mixing trend, displaying lower Se/S relative to the δ34S values.
from group 2 show a wider range of values to lower Se/S This may reflect lower temperatures, since the fluid-sulfide
but at relatively constant δ34S, although sample AD0093-14 Se partition coefficient has been proposed to depend on tem-
from the dacite dome, together with group 3 samples (in perature (Fitzpatrick, 2008). As sample DH996-2 comes from
situ and bulk measurements), has higher δ34S values and the upper part of the Santos deposit, it is possible that system
lower Se/S ratios compared with the majority of the pyrite was cooling to lower temperatures in this area.
analyses (Fig. 14). Results from these samples are consis- The Candelaria-Punta del Cobre district has distinct geo-
tent with a different fluid potentially sourced from or equili- logic features that may be particularly favorable for IOCG
brated with basinal rocks. A generalized mixing trend based mineralization. The important stratigraphically controlled
on previously defined deposition conditions for Se/S (Fitz- manto mineralization is hosted by the volcanic-sedimentary
patrick, 2008) is suggested on Figure 14 with end-member unit in the upper part of the Punta del Cobre Formation,
magmatic-hydrothermal and basin-derived fluids. A similar below, and locally within, the sedimentary sequence at the
mixing trend was interpreted for the IOCG mineralization base of the Chañarcillo Group. The volcanic-sedimentary
at Mantoverde and Mina Justa IOCG deposits (Benavides unit provided a zone of high permeability that may have been
et al., 2007; Fitzpatrick, 2008; Li et al., 2018). Additionally, accessed by deep fluids and also by fluids from the Chañar-
recent δ18O data obtained from magnetite samples within cillo Group facilitating sulfide precipitation.
the manto horizon in the Candelaria deposit indicate a mix- Las Pintadas is located at the southern end of the district.
ture between magmatic-hydrothermal and external fluids Sample LP-1 from Las Pintadas deposit has δ34S values and
(Rodriguez-Mustafá et al., in press). The most positive δ34S Se/S and Ni/Se ratios that are consistent with the rest of the

Magmac-derived
~-2-2.4‰ ; Se/S ~ 1.2-5*10-4
LP-1

DH996-23 ES032-15
DH996-21
AD0093-14
AD006-26
0.0001
mixi
ng tr
ES032-5 end
Se/S

DH996-2
LD1493-9
AD357-3

AD357-8
AD357-10 AD009-14

Basinal, Seawater derived

0.00001 ~+21‰ ; Se/S ~ 5-25*10-8

-2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12
δ34S(‰ )
Group 1 Group 2 Group 3 Group 4 Pyrite type
Structure hosted Manto hosted LD1493-9 Rim
DH996-2 LP-1
DH996-23 ES032-15 AD357-3 Core
DH996-21 AD006-26 ES032-5 Grain aggregates
AD357-10
AD0093-14 Whole grain
AD357-8
AD009-14
Fig. 14. Pyrite chemistry integrated with δ34S measurements on pyrite: Se/S and δ34S used for evaluating fluid source; higher
δ34S and lower Se/S correlate with seawater (or basin)-derived fluids, lower δ34S and higher Se/S correlate with a magmatic-
derived fluid. Source ranges were obtained by Fitzpatrick (2008).
22 DEL REAL ET AL.

sample set, suggesting that fluids and processes were generally Acknowledgments
similar to those in the main part of the district, even though Lundin Mining is acknowledged and thanked for funding
the mineralization is at higher stratigraphic levels than the field support and analytical results. Sulfur isotope analysis
main cluster of IOCG deposits (Fig. 2). As discussed previ- were partially funded by National Science Foundation (NSF)
ously, pyrite in sample LP-1 also has distinct chemical differ- grant 1524394 to Adam Simon. Martin Reich thanks support
ences from the rest of the sample set, with poor correlations from Millennium Science Initiative through the grant “Mil-
between As and Co and Se and Ni, well-correlated Co and Se, lennium Nucleus for Metal Tracing Along Subduction.” We
and higher Co and Se contents. The role of different fluids, thank the support of Louisa Smieska and Rong Huang from
stratigraphic position, and wall-rock compositions or the distal the Cornell High Energy Synchrotron source, Jay Thomas
location of Las Pintadas relative the core of the Candelaria- and William Nachlas from the Syracuse University EPMA
Punta del Cobre system may have played a role in controlling laboratory, Daniel Layton-Matthews and Alexandre Voinot
these differences. Further work would be required to evalu- from the Queens Facility for Isotope Research, and finally
ate these options. Kouki Kitajima, John Valley, and Michael Spicuzza from the
Conclusions WiscSIMS laboratory of the University of Madison Wiscon-
sin; WiscSIMS is supported by NSF-EAR1658823 and the
Pyrite in samples from IOCG mineralization in several loca- University of Wisconsin-Madison. Lluis Fontboté and Lucas
tions and stratigraphic levels within the Candelaria-Punta Marshall are thanked for thorough reviews that significantly
del Cobre district shows distinct variations of Ni, Co, and improved the paper.
Se concentrations that are interpreted to reflect changes in
temperature, redox, source of the hydrothermal fluid(s), and REFERENCES
potentially stratigraphic/host-rock controls. The δ34S isotope Abraitis, P.K., Pattrick, R.A.D., and Vaughan, D.J., 2004, Variations in the
values show a range of values (~–2 to 12‰) where most of compositional, textural and electrical properties of natural pyrite: A review:
the data are concentrated between –2 and 2‰. The pres- International Journal of Mineral Processing, v. 74, p. 41–59.
ence of varying Fe oxides (magnetite, partly as mushketovite Arabasz, W., 1971, Geological and geophysical studies of the Atacama fault
and hematite) in the system suggests that changes in the δ34S zone in northern Chile: Ph.D. thesis, Pasadena, California, California Insti-
tute of Technology, 264 p.
values may correlate with changes in the redox state of the Arévalo, C., 1999, The Coastal Cordillera/Precordillera boundary in the
hydrothermal fluid. This may partially explain changes in Tierra Amarilla area (27° 20' 27 40'S/70° 05'–70 20'W), northern Chile, and
Ni/Se ratios observed in some of the pyrite samples. the structural setting of the Candelaria Cu-Au ore deposit: Unpublished
The Co/Ni ratios combined with Se/S and δ34S suggest that Ph.D. thesis, Kingston, Canada, University of Queens.
Arévalo, C., Grocott, J., Martin, W., Pringle, M., and Taylor, G., 2006, Struc-
the fluids responsible for the main mineralization in the dis- tural setting of the Candelaria Fe oxide Cu-Au deposit, Chilean Andes
trict have a magmatic-hydrothermal origin. High δ34S values (27°30'S): Economic Geology, v. 101, p. 819–841.
in pyrite from late veins suggest that an externally derived Bajwah, Z.U., Seccombe, P.K., and Offler, R., 1987, Trace element distribu-
fluid entered the system toward the end of the main mineral- tion, Co:Ni ratios and genesis of the Big Cadia iron-copper deposit, New
izing event. This fluid is interpreted to be of basinal origin, South Wales, Australia: Mineralium Deposita, v. 22, p. 292–300.
Barker, S.L.L., Hickey, K.A., Cline, J.S., Dipple, G.M., Kilburn, M.R.,
possibly derived from the overlying Chañarcillo Group. Vaughan, J.R., and Longo, A.A., 2009, Uncloaking invisible gold: Use of
This study demonstrates that pyrite chemistry combined nanoSIMS to evaluate gold, trace elements, and sulfur isotopes in pyrite
with in situ δ34S analyses provides useful information on from Carlin-type gold deposits: Economic Geology, v. 104, p. 897–904.
the fluid sources for IOCG deposits. As outlined, the fluid Barra, F., Reich, M., Selby, D., Rojas, P., Simon, A., Salazar, E., and Palma,
source and nature of IOCG mineralization has been a topic of G., 2017, Unraveling the origin of the Andean IOCG clan: A Re-Os isotope
approach: Ore Geology Reviews, v. 81, p. 62–78.
debate. Our results concur with recent results from the Mina Barton, M.D., and Johnson, D.A., 1996, Evaporitic-source model for igneous-
Justa IOCG deposit in Southern Peru (Li et al., 2018), where related Fe oxide-(REE-Cu-Au- U) mineralization: Geology, v. 24, p. 259–262.
mineralization is interpreted to be of magmatic-hydrothermal Benavides, J., Kyser, T.K., Clark, A.H., Oates, C.J., Zamora, R., Tarnovschi,
origin with a late incursion of basin-derived fluids. The results R., and Castillo, B., 2007, The Mantoverde iron oxide-copper-gold district,
III Región, Chile: The role of regionally derived, nonmagmatic fluids in
from these deposits do not support formation of IOCG miner- chalcopyrite mineralization: Economic Geology, v. 102, p. 415–440.
alization solely from oxidized saline brines (Barton and John- Bilenker, L.D., Simon, A.C., Reich, M., Lundstrom, C.C., Gajos, N., Bin-
son, 1996), at least in the Andean belt. deman, I., Barra, F., and Munizaga, R., 2016, Fe-O stable isotope pairs
The Co/Ni ratios and elevated Co and Ni concentrations elucidate a high-temperature origin of Chilean iron oxide-apatite deposits:
obtained from the pyrite samples in the district suggest that Geochimica et Cosmochimica Acta, v. 177, p. 94–104.
Boric, R., Díaz, F., and Maksaev, V., 1990, Geología y yacimientos metalífe-
at least part of the hydrothermal fluid has a mafic igneous ros de la Región de Antofagasta: Servicio Nacional de Geología y Minería,
affinity, as has been proposed by Chiaradia et al. (2006) on Boletín 40, 246 p.
the basis of δ37Cl isotope data. Mineralization associated with Bralia, A., Sabatini, G., and Troja, F., 1979, A revaluation of the Co/Ni ratio in
magmatic fluids of a mafic affinity has been proposed to result pyrite as geochemical tool in ore genesis problems: Mineralium Deposita,
v. 14, p. 353–374.
from hydrothermal activity driven by mantle underplating Campbell, F.A., and Ethier, V.G., 1984, Nickel and colbalt in pyrrhotite and
(Groves et al., 2010). In the case of the Andean IOCG belt, pyrite from the Faro and Sullivan ore bodies: The Canadian Mineralogist,
mineralization is coeval with or immediately follows a long v. 22, p. 503–506.
period of subduction-related extension, which could have Chambers, L.A., 1982, Sulfur isotope study of a modern intertidal environ-
facilitated back-arc asthenospheric upwelling (Mpodozis and ment, and the interpretation of ancient sulfides: Geochimica et Cosmochi-
mica Acta, v. 46, p. 721–728.
Ramos, 1989). This setting is consistent with a model of min- Chen, H., Kyser, T.K., and Clark, A.H., 2011, Contrasting fluids and reser-
eralization associated with fluids sourced from mafic, prob- voirs in the contiguous Marcona and Mina Justa iron oxide-Cu (-Ag-Au)
ably mantle-derived magmas. deposits, south-central Perú: Mineralium Deposita, v. 46, p. 677–706.
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 23

Chen, H., Cooke, D.R., and Baker, M.J., 2013, Mesozoic iron oxide copper- S.W., 2015, Trace element content of sedimentary pyrite in black shales:
gold mineralization in the central Andes and the Gondwana Supercontinent Economic Geology, v. 110, p. 1389–1410.
breakup: Economic Geology, v. 108, p. 37–44. Griffin, W.L., Ashley, P.M., Ryan, C.G., Sie, S.H., and Suter, G.F., 1991, Pyrite
Chiaradia, M., Banks, D., Cliff, R., Marschik, R., and Haller, A., 2006, Origin geochemistry in the North Arm epithermal Ag-Au deposit, Queensland,
of fluids in iron oxide-copper-gold deposits: Constraints from 37Cl, 87Sr/86Sr Australia. A proton-microprobe study: The Canadian Mineralogist, v. 29,
and Cl/Br: Mineralium Deposita, v. 41, p. 565–573. p. 185–198.
Claypool, G.E., Holser, W.T., Kaplan, I.R., Sakai, H., and Zak, I., 1980, The Grocott, J., and Taylor, G.K., 2002, Magmatic arc fault systems, deformation
age curves of sulfur and oxygen isotopes in marine sulfate and their mutual partitioning and emplacement of granitic complexes in the Coastal Cordil-
interpretation: Chemical Geology, v. 28, p. 199–260. lera, north Chilean Andes (25 30'S to 27 00'S): Journal of the Geological
Cook, N., and Chryssoulis, S.L., 1990, Concentrations of “invisible gold” in Society of London, v. 159, p. 425–443.
the common sulfides: The Canadian Mineralogist, v. 28, p. 1–16. Grocott, J., Brown, M., Dallmeyer, R.D., Taylor, G.K., and Treloar, P.J., 1994,
Cook, N.J., Ciobanu, C.L., and Mao, J., 2009, Textural control on gold dis- Mechanisms of continental growth in extensional arcs: An example from the
tribution in As-free pyrite from the Dongping, Huangtuliang and Hougou Andean plate-boundary zone: Geology, v. 22, p. 391–394.
gold deposits, North China craton (Hebei Province, China): Chemical Groves, D.I., Bierlein, F.P., Meinert, L.D., and Hitzman, M.W., 2010, Iron
Geology, v. 264, p. 101–121. oxide copper-gold (IOCG) deposits through earth history: Implications for
Corriveau, L., Montreuil, J.F., and Potter, E.G., 2016, Alteration facies origin, lithospheric setting, and distinction from other epigenetic iron oxide
linkages among iron oxide copper-gold, iron oxide-apatite, and affiliated deposits: Economic Geology, v. 105, p. 641–654.
deposits in the Great Bear magmatic zone, Northwest Territories, Canada: Gülaçar, O.F., and Delaloye, M., 1976, Geochemistry of nickel, cobalt and
Economic Geology, v. 111, p. 2045–2072. copper in alpine-type ultramafic rocks: Chemical Geology, v. 17, p. 269–280.
Crowe, D.E., and Vaughan, R.G., 1996, Characterization and use of isotopi- Hannington, M.D., Poulsen, K.H., Thompson, J.F.H., and Sillitoe, R.H.,
cally homogeneous standards for in situ laser microprobe analysis of 34S/32S 1999, Volcanogenic gold in the massive sulfide environment: Reviews in
ratios: American Mineralogist, v. 81, p. 187–193. Economic Geology, v. 8, p. 325–356.
Dale, D., 2015, Praxes software: https://github.com/praxes/praxes. Haynes, D.W., Cross, K.C., Bills, R.T., and Reed, M.H., 1995, Olympic Dam
Deditius, A.P., Utsunomiya, S., Ewig, R.C., Chryssoulis, S.L., Venter, D., and ore genesis; a fluid mixing model: Economic Geology, v. 90, p. 281–307.
Kesler, S.E., 2009, Decoupled geochemical behavior of As and Cu in hydro- Hedenquist, J.W., and Lowenstern, J.B., 1994, The role of magmas in the
thermal systems: Geology, v. 37, p. 707–710. formation of hydrothermal ore deposits: Nature, v. 370, p. 519–527.
Deditius, A.P., Kesler, S.E., Ewing, R.C., and Walshe, J., 2011, Trace metal Heinrich, C.A., Halter, W., Landtwing, M.R., and Pettke, T., 2008, The for-
nanoparticles in pyrite: Ore Geology Reviews, v. 42, p. 32–46. mation of economic porphyry copper (-gold) deposits: Constraints from
Deditius, A.P., Reich, M., Kesler, S.E., Utsunomiya, S., Chryssoulis, S.L., microanalysis of fluid and melt inclusions: Geological Society, London,
Walshe, J., and Ewing, R.C., 2014, The coupled geochemistry of Au and Special Publications, v. 248, p. 247–263.
As in pyrite from hydrothermal ore deposits: Geochimica et Cosmochimica Helgeson, H.C., Delany, J.M., Nesbitt, H.W., and Bird, D.K., 1978, Summary
Acta, v. 140, p. 644–670. and critique of the thermodynamic properties of rock-forming minerals:
De Haller, A., and Fontboté, L., 2009, The Raul-Condestable iron oxide cop- American Journal of Science, v. 178, p. 1–229.
per-gold deposit, central coast of Peru: Ore and related hydrothermal alter- Hitzman, M., 2000, Iron oxide-Cu-Au deposits: What, where, when, and why:
ation, sulfur isotopes, and thermodynamic constraints: Economic Geology, Deposits: A global perspective, in Porter, T.M., ed., Hydrothermal iron
v. 104, p. 365–384.
oxide copper gold and related deposits: A global perspective, v. 1: Adelaide,
De Haller, A., Corfu, F., Fontboté, L., Schaltegger, U., Barra, F., Chiara-
Australian Mineral Foundation, p. 9–25.
dia, M., Frank, M., and Alvarado, J.Z., 2006, Geology, geochronology, and
Hitzman, M., Oreskes, N., and Einaudi, M., 1992, Geological characteristics
Hf and Pb isotope data of the Raúl-Condestable iron oxide-copper-gold
and tectonic setting of proterozoic iron oxide (Cu-U-Au-REE) deposits:
deposit, central coast of Peru: Economic Geology, v. 101, p. 281–310.
Precambrian Research, v. 58, p. 241–287.
del Real, I., and Arriagada, C., 2015, Inversión tectónica positiva en el distrito
Hopf, S., 1987, Petrographische, mineralogische und geochemische Beobach-
El Espino: Relaciones entre deformación, magmatismo y mineralizacion
tungen an der Cu-Lagerstätte Agustina/Distrikt Punta del Cobre/Chile:
IOCG, Provincia de Choapa, Chile: XIV Congreso Geológico Chileno, La
Unpublished Diplomarbeit, Heidelberg, Universität Heidelberg, 144 p.
Serena, 2015, Proceedings, 4 p.
Huston, D.L., Sie, S.H., Suter, G.F., Cooke, D.R., and Both, R.A., 1995a,
del Real, I., Thompson, J.F.H., and Carriedo, J., 2018, Lithological and
structural controls on the genesis of the Candelaria-Punta del Cobre iron Trace elements in sulfide minerals from eastern Australian volcanic-hosted
oxide copper gold district, northern Chile: Ore Geology Reviews, v. 102, massive sulfide deposits: Part I. Proton microprobe analyses of pyrite, chal-
p. 106–153. copyrite, and sphalerite, and Part II. Selenium levels in pyrite; comparison
del Real, I., Smieska, L., Thompson, J.F.H., Martinez, C., Thomas, J., and with δ34S values and implications for the source of sulfure in volcanogenic
Layton-Matthews, D., 2019, Using multiple micro-analytical techniques for hydrothermal systems: Economic Geology, v. 90, p. 1167–1196.
evaluating quantitative synchrotron-XRF elemental mapping of hydrother- Huston, D.L., Sie, S.H., and Suter, G.F., 1995b, Selenium and its impor-
mal pyrite: Journal of Analytical and Atomic Spectroscopy, doi: 10.1039/ tance to the study of ore genesis: The theoretical basis and its applica-
C9JA00083F. tion to volcanic-hosted massive sulfide deposits using pixeprobe analysis:
Einaudi, M.T., Hedenquist, J.W., and Inan, E.E., 2003, Sulfidation state of Nuclear Instruments and Methods in Physics Research, Section B, v. 104,
fluids in active and extinct hydrothermal systems: Transitions from por- p. 476–480.
phyry to epithermal environments: Society of Economic Geologists, Special Jamtveit, B., 1991, Oscillatory zonation patterns in hydrothermal grossular-
Publication 10, p. 285–314. andradite garnet: Nonlinear dynamics in regions of immiscibility: American
Espinoza, S., Véliz, H., Esquivel, J., Arias, J., and Moraga, A., 1996, The Mineralogist, v. 76, p. 1319–1327.
cupriferous province of the coastal range, northern Chile: Andean Copper Jenner, F.E., O’Neill, H.S.C., Arculus, R.J., and Mavrogenes, J.A., 2010, The
Deposits, v. 5, p. 19–32. magnetite crisis in the evolution of arc-related magmas and the initial con-
Fitzpatrick, A.J., 2008, The measurement of the Se/S ratios in sulphide min- centration of Au, Ag and Cu: Journal of Petrology, v. 51, p. 2445–2464.
erals and their application to ore deposit studies: Ph.D. disseration, Kings- Johnson, T.M., 2004, A review of mass-dependent fractionation of selenium
ton, Ontario, Canada, Queen’s University, 188 p. isotopes and implications for other heavy stable isotopes: Chemical Geol-
Franchini, M., McFarlane, C., Maydagán, L., Reich, M., Lentz, D.R., Mein- ogy, v. 204, p. 201–214.
ert, L., and Bouhier, V., 2015, Trace metals in pyrite and marcasite from Jonsson, E., Troll, V.R., Högdahl, K., Harris, C., Weis, F., Nilsson, K.P., and
the Agua Rica porphyry-high sulfidation epithermal deposit, Catamarca, Skelton, A., 2013, Magmatic origin of giant “Kiruna-type” apatite-iron-
Argentina: Textural features and metal zoning at the porphyry to epithermal oxide ores in central Sweden: Scientific Reports, v. 3, article 1644.
transition: Ore Geology Reviews, v. 66, p. 366–387. Kajiwara, Y., and Krouse, H.R., 1971, Sulfur isotope partitioning in metallic
Giggenbach, W.F., 1997, The origin and evolution of fluids in magmatic- sulfide systems: Canadian Journal of Earth Sciences, v. 8, p. 1397–1408.
hydrothermal systems, in Barnes, H.L., ed., Geochemistry of hydrothermal Keith, M., Haase, K.M., Klemd, R., Krumm, S., and Strauss, H., 2016, Sys-
ore deposits, 3rd ed.: New York, Wiley, p. 737–796. tematic variations of trace element and sulfur isotope compositions in pyrite
Gregory, D.D., Large, R.R., Halpin, J.A., Baturina, E.L., Lyons, T.W., Wu, with stratigraphic depth in the Skouriotissa volcanic-hosted massive sulfide
S., Danyushevsky, L., Sack, P.J., Chappaz, A., Maslennikov, V.V., and Bull, deposit, Troodos ophiolite, Cyprus: Chemical Geology, v. 423, p. 7–18.
24 DEL REAL ET AL.

Keith, M., Smith, D.J., Jenkin, G.R.T., Holwell, D.A., and Dye, M.D., 2017, oxide-copper-gold deposit, Carajás mineral province, Brazil: Mineralium
A review of Te and Se systematics in hydrothermal pyrite from precious Deposita, v. 43, p. 129–159.
metal deposits: Insights into ore-forming processes: Ore Geology Reviews, Mpodozis, C., and Ramos, V.A., 1989, The Andes of Chile and Argentina:
v. 96, p. 269–282. Geology of the Andes and its relation to hydrocarbon and mineral resources:
Knipping, J.L., Bilenker, L.D., Simon, A.C., Reich, M., Barra, F., Deditius, Circum-Pacific Council for Energy and Mineral Resources, Earth Science
A.P., Lundstrom, C., Bindeman, I., and Munizaga, R., 2015a, Giant Kiruna- Series, v. 11, p. 59–90.
type deposits form by efficient flotation of magmatic magnetite suspen- Myers, J.T., and Eugster, H.P., 1983, The system Fe-Si-O: Oxygen buffer
sions: Geology, v. 43, p. 591–594. calibrations to 1,500 K: Contributions to Mineralogy and Petrology, v. 82,
Knipping, J.L., Bilenker, L.D., Simon, A.C., Reich, M., Barra, F., Deditius, p.75–90.
A.P., Wälle, M., Heinrich, C.A., Holtz, F., and Munizaga, R., 2015b, Trace Nadeau, O., Williams-Jones, A.E., and Stix, J., 2010, Sulphide magma as a
elements in magnetite from massive iron oxide-apatite deposits indicate a source of metals in arc-related magmatic hydrothermal ore fluids: Nature
combined formation by igneous and magmatic-hydrothermal processes: Geoscience, v. 3, p. 501–506.
Geochimica et Cosmochimica Acta, v. 171, p. 15–38. Nicholls, I.A., Whitford, D.J., Harris, K.L., and Taylor, S.R., 1980, Variation
Koglin, N., Frimmel, H.E., Minter, W.E.L., and Brätz, H., 2010, Trace-ele- in the geochemistry of mantle sources for tholeiitic and calc-alkaline mafic
ment characteristics of different pyrite types in Mesoarchaean to Palaeo- magmas, Western Sunda volcanic arc, Indonesia: Chemical Geology, v. 30,
proterozoic placer deposits: Mineralium Deposita, v. 45, p. 259–280. p. 177–199.
Kozdon, R., Kita, N.T., Huberty, J.M., Fournelle, J.H., Johnson, C.A., and Ohmoto, H., 1972, Systematics of sulfur and carbon isotopes in hydrothermal
Valley, J.W., 2010, In situ sulfur isotope analysis of sulfide minerals by ore deposits: Economic Geology, v. 67, p. 551–578.
SIMS: Precision and accuracy, with application to thermometry of ~3.5 Ga ——2003, Nonredox transformations of magnetite-hematite in hydrothermal
Pilbara cherts: Chemical Geology, v. 275, p. 243–253. systems: Economic Geology, v. 98, p. 157–161.
Kress, V.C., and Carmichael, I.S.E., 1991, The compressibility of silicate liq- Ohmoto, H., and Goldhaber, M., 1997, Sulfur and carbon isotopes, in Barnes,
uids containing Fe2O3 and the effect of composition, temperature, oxygen H.L., ed., Geochemistry of hydrothermal ore deposits: Canada, Wiley,
fugacity and pressure on their redox states: Contributions to Mineralogy p. 517–600.
and Petrology, v. 108, p. 82–92. Ohmoto, H., and Lasaga, A.C., 1982, Kinetics of reaetions between aqueous
Krumm, S., Keith, M., Klemd, R., Strauss, H., and Haase, K.M., 2015, Sys- sulfates and sulfides in hydrothermal systems: Geochemica et Cosmochi-
tematic variations of trace element and sulfur isotope compositions in pyrite mica Acta, v. 46, p. 6141–6156.
with stratigraphic depth in the Skouriotissa volcanic-hosted massive sulfide Otake, T., Wesolowski, D.J., Anovitz, L.M., Allard, L.F., and Ohmoto, H.,
deposit, Troodos ophiolite, Cyprus: Chemical Geology, v. 423, p. 7–18. 2010, Mechanisms of iron oxide transformations in hydrothermal systems:
Large, R.R., Danyushevsky, L., Hollit, C., Maslennikov, V., Meffre, S., Gil- Geochimica et Cosmochimica Acta, v. 74, p. 6141–6156.
bert, S., Bull, S., Scott, R., Emsbo, P., Thomas, H., Singh, B., and Foster, Peterson, E., and Mavrogenes, J., 2014, Linking high-grade gold mineral-
J., 2009, Gold and trace element zonation in pyrite using a laser imaging ization to earthquake-induced fault-valve processes in the Porguera gold
technique: Implications for the timing of gold in orogenic and Carlin-style deposit, Papua New Guinea: Geology, v. 42, p. 383–386.
sediment-hosted deposits: Economic Geology, v. 104, p. 635–668. Pollard, P.J., 2006, An intrusion-related origin for Cu-Au mineralization in
Large, R.R., Halpin, J.A., Danyushevsky, L.V., Maslennikov, V.V., Bull, S.W., iron oxide-copper-gold (IOCG) provinces: Mineralium Deposita, v. 41,
Long, J.A., Gregory, D.D., Lounejeva, E., Lyons, T.W., Sack, P.J., McGold- p. 179–187.
rick, P.J., and Calver, C.R., 2014, Trace element content of sedimentary Putnis, A., Fernandez-Diaz, L., and Prieto, M., 1992, Experimentally pro-
pyrite as a new proxy for deep-time ocean-atmosphere evolution: Earth and duced oscillatory zoning in the (Ba, Sr)SO4 solid solution: Nature, v. 358,
Planetary Science Letters, v. 389, p. 209–220. p. 743–745.
Lehner, S.W., Savage, K.S., and Ayers, J.C., 2006, Vapor growth and charac- Reich, M., Kesler, S.E., Utsunomiya, S., Palenik, C.S., Chryssoulis, S.L., and
terization of pyrite (FeS2) doped with Co, Ni, and As: Variations in semicon- Ewing, R.C., 2005, Solubility of gold in arsenian pyrite: Geochimica et Cos-
ducting properties: Journal of Crystal Growth, v. 286, p. 306–317. mochimica Acta, v. 69, p. 2781–2796.
Li, R., Chen, H., Xia, X., Yang, Q., Danyushevsky, L.V., and Lai, C., 2018, Reich, M., Deditius, A., Chryssoulis, S., Li, J.-W., Ma, C.-Q., Parada, M.A.,
Using integrated in-situ sulfide trace element geochemistry and sulfur Barra, F., and Mittermayr, F., 2013, Pyrite as a record of hydrothermal fluid
isotopes to trace ore-forming fluids: Example from the Mina Justa IOCG evolution in a porphyry copper system: A SIMS/EMPA trace element study:
deposit (southern Perú): Ore Geology Reviews, v. 101, p. 165–179. Geochimica et Cosmochimica Acta, v. 104, p. 42–62.
Liang, H.Y., Sun, W., Su, W.C., and Zartman, R.E., 2009, Porphyry copper- Reich, M., Simon, A., Deditius, A., Barra, F., Chryssoulis, S., et al., 2016,
gold mineralization at Yulong, China, promoted by decreasing redox poten- Trace element signature of pyrite from the Los Colorados iron oxide-apatite
tial during magnetite alteration: Economic Geology, v. 104, p. 587–596. (IOA) deposit, Chile: A missing link between Andean IOA and iron oxide
Longhi, J., Durand, S.R., and Walker, D., 2010, The pattern of Ni and Co copper-gold systems?: Economic Geology, v. 11, p. 743–761.
abundances in lunar olivines: Geochimica et Cosmochimica Acta, v. 74, Requia, K., and Fontboté, L., 1999, Hydrothermal alkali metasomatism in
p. 784–798. the Salobo iron oxide Cu (-Au) deposit, Carajás mineral province, north-
Lopez, G.P., Hitzman, M.W., and Nelson, E.P., 2014, Alteration patterns and ern Brazil: Society for Geology Applied to Mineral Deposits (SGA) Bien-
structural controls of the El Espino IOCG mining district, Chile: Minera- nial Meeting, 5th, London, England, August 22–25, 1999, Proceedings,
lium Deposita, v. 49, p. 235–259. p. 1025–1028.
Marschik, R., and Chiaradia, M., 2000, Lead isotope signatures of ore, volca- Revan, M.K., Genç, Y., Maslennikov, V.V., Maslennikova, S.P., Large, R.R.,
nic, and batholith rocks of the Candelaria-Punta del Cobre area: Interna- and Danyushevsky, L.V., 2014, Mineralogy and trace-element geochemistry
tional Geological Congress, 31st, Rio de Janeiro, Brazil, August 6–17, 2000, of sulfide minerals in hydrothermal chimneys from the Upper-Cretaceous
Abstracts, CD-ROM. VMS deposits of the eastern Pontide orogenic belt (NE Turkey): Ore Geol-
Marschik, R., and Fontboté, L., 2001a, The Punta del Cobre Formation, ogy Reviews, v. 63, p. 129–149.
Punta del Cobre- Candelaria area, northern Chile: Journal of South Ameri- Richards, J.P., Lopez, G.P., Zhu, J.J., Creaser, R.A., Locock, A.J., and Mumin,
can Earth Sciences, v. 14, p. 401–433. A.H., 2017, Contrasting tectonic settings and sulfur contents of magmas
——2001b, The Candelaria-Punta del Cobre iron oxide Cu-Au (-Zn-Ag) associated with Cretaceous porphyry Cu ± Mo ± Au and intrusion-related
deposits, Chile: Economic Geology, v. 96, p. 1799–1826. iron oxide Cu-Au deposits in northern Chile: Economic Geology, v. 112,
Marschik, R., and Söllner, F., 2006, Early Cretaceous U-Pb zircon ages for p. 295–318.
the Copiapo plutonic complex and implications for the IOCG mineraliza- Roberts, D.E., and Hudson, G.R.T., 1983, The Olympic Dam copper-ura-
tion at Candelaria, Atacama region, Chile: Mineralium Deposita, v. 41, nium-gold deposit, Roxby Downs, South Australia: Economic Geology,
p. 785–801. v. 78, p. 799–822.
Mathur, R., Marschik, R., Ruiz, J., Munizaga, F., Leveille, R., and Martin, Rodriguez-Mustafá, M., Simon, A.C., del Real, I., Thompson, J.F.H., Bilen-
W., 2002, Age of mineralization of the Candelaria Fe oxide Cu-Au deposit ker, L., Barra F., and Bindeman I., in press, A continuum from iron oxide
and the origin of the Chilean iron belt, based on Re-Os isotopes: Economic copper-gold to iron oxide-apatite deposits: Evidence from Fe and O stable
Geology, v. 97, p. 59–71. isotopes and trace element chemistry of magnetite: Economic Geology.
Monteiro, L., Xavier, R., and Carvalho, E. de, 2008, Spatial and temporal Rojas, P. A., Barra, F., Reich, M., Deditius, A., Simon, A., Uribe, F., Romero,
zoning of hydrothermal alteration and mineralization in the Sossego iron R., and Rojo, M., 2018, A genetic link between magnetite mineralization
 CANDELARIA-PUNTA DEL COBRE IOCG DISTRICT, CHILE 25

and diorite intrusion at the El Romeral iron oxide-apatite deposit, northern Taylor, S.R., Kaye, M., White, A.J.R., Duncan, A.R., and Ewart, A., 1969,
Chile: Mineralium Deposita, v. 53, p. 947–966. Genetic significance of Co, Cr, Ni, Sc and V content of andesites: Geochi-
Román, N., Reich, M., Leisen, M., Morata, D., Barra, F., and Deditius, A.P., mica et Cosmochimica Acta, v. 33, p. 275–286.
2019, Geochemical and micro-textural fingerprints of boiling in pyrite: Thomson, J., Higgs, N.C., Croudace, I.W., Colley, S., and Hydes, D.J., 1993,
Geochimica et Cosmochimica Acta, v. 246, p. 60–85. Redox zonation of elements at an oxic/post-oxic boundary in deep-sea sedi-
Rudnick, R.L., and Taylor, S.R., 1987, The composition and petrogenesis of ments: Geochimica et Cosmochimica Acta, v. 57, p. 579–595.
the lower crust: A xenolith study: Journal of Geophysical Research: Solid Tossell, J.A., Vaughan, D.J., and Burdett, J.K., 1981, Pyrite, marcasite, and
Earth, v. 92, p. 13981–14005. arsenopyrite type minerals: Crystal chemical and structural principles:
Rusk, B., Oliver, N., Cleverley, J., Blenkinsop, T., and Zhang, D., 2010, Physi- Physics and Chemistry of Minerals, v. 7, p. 177–184.
cal and chemical characteristics of the Ernest Henry iron oxide copper gold Ueda, A., and Sakai, H., 1984, Sulfur isotope study of Quaternary volcanic
deposit, Australia; implications for IOGC genesis, in Porter, T.M., Hydro- rocks from the Japanese Islands arc: Geochimica et Cosmochimica Acta,
thermal iron oxid copper-gold and related deposits: A global perspective: v. 48, p. 1837–1848.
Advances in the understanding of IOCG deposits: Linden Park, South Aus- Ullrich, T.D., and Clark, A.H., 1999, The Candelaria Cu/Au deposit, III
tralia, PGC Publishing, Global Perspective Series, v. 3, p. 201–218. Región, Chile: Paragenesis, geochronology and fluid composition, in
Rye, R.O., and Ohmoto, H., 1974, Sulfur and carbon isotopes and ore gen- Stanley, C.J., ed., Mineral deposits: Processes to processing: Rotterdam,
esis: A review: Economic Geology, v. 69, p. 826–842. Balkema, p. 201–204.
Rye, R.O., Luhr, J.F., and Wasserman, M.D., 1984, Sulfur and oxygen isoto- Ushikubo, T., Williford, K.H., Farquhar, J., Johnston, D.T., Van Kranendonk,
pic systematics of the 1982 eruptions of El Chichón volcano, Chiapas, Mex- M.J., and Valley, J.W., 2014, Development of in situ sulfur four-isotope
ico: Journal of Volcanology and Geothermal Research, v. 23, p. 109–123. analysis with multiple Faraday cup detectors by SIMS and application to
Salazar, E., Barra, F., Reich, M., Simon, A., Leisen, M., Palma, G., Romero, pyrite grains in a Paleoproterozoic glaciogenic sandstone: Chemical Geol-
R., and Rojo, M., 2019, Trace element geochemistry of magnetite from the ogy, v. 383, p. 86–99.
Cerro Negro Norte iron oxide-apatite deposit, northern Chile: Mineralium Williams, P.J., Barton, M.D., Johnson, D.A., Fontboté, L., De Haller, A.,
Deposita, v. 55, p. 1–20 Mark, G., Oliver, N.H.S., and Marshick, R., 2005, Iron oxide copper-gold
Seal, R.R., 2006, Sulfur isotope geochemistry of sulfide minerals: Reviews in deposits: Geology, space-time distribution, and possible modes of origin:
Mineralogy and Geochemistry, v. 61, p. 633–677. Economic Geology 100th Anniversary Volume, p. 371–405.
Sillitoe, R.H., 2003, Iron oxide-copper-gold deposits: An Andean view: Min- Zhao, H.-X., Frimmel, H.E., Jiang, S.-Y., and Dai, B.-Z., 2011, LA-ICP-MS
eralium Deposita, v. 38, p. 787–812. trace element analysis of pyrite from the Xiaoqinling gold district, China:
Simon, A.C., Knipping, J., Reich, M., Barra, F., Deditius, A.P., Bilenker, L., Implications for ore genesis: Ore Geology Reviews, v. 43, p. 142–153.
and Childress, T., 2018, Kiruna-type iron oxide-apatite ( IOA ) and iron Zhao, X.-F., and Zhou, M.-F., 2011, Fe-Cu deposits in the Kangdian region,
oxide copper-gold (IOCG ) deposits form by a combination of igneous and SW China: A Proterozoic IOCG (iron-oxide–copper–gold) metallogenic
magmatic-hydrothermal processes: Evidence from the Chilean iron belt: province: Mineralium Deposita, v. 46, p. 731–747.
Society of Economic Geologists, Special Publication 21, p. 89–114.
Solé, V.A., Papillon, E., Cotte, M., Walter, P., and Susini, J., 2007, A multiplat-
form code for the analysis of energy-dispersive X-ray fluorescence spectra:
Spectrochimica Acta —Part B Atomic Spectroscopy, v. 62, p. 63–68.
Strauss, H., 1997, The isotopic composition of sedimentary sulfur through
time: Palaeogeography Paleaoclimatology Palaeoecology, v. 132, p. 97–118.
——1999, Geological evolution from isotope proxy signals—sulfur: Chemical
Geology, v. 161, p. 89–101.
Sun, W., Arculus, R.J., Kamenetsky, V.S., and Binns, R.A., 2004, Release of
gold-bearing fluids in convergence margin magmas prompted by magnet-
ide crystallization: Nature, v. 431, p. 975–978.
Sun, W., Liang, H., Ling, M., Zhan, M., Ding, X., Zhang, H., Yang, X., Li, Y., Irene del Real earned her degree in geology from
Ireland, T.R., Wei, Q., and Fan, W., 2013, The link between reduced por- the University of Chile, an M.Sc. degree from the
phyry copper deposits and oxidized magmas: Geochimica et Cosmochimica University of British Columbia, Canada, and a
Acta, v. 103, p. 263–275. Ph.D. degree from Cornell University, USA. Hav-
Tanner, D., Henley, R.W., Mavrogenes, J.A., and Holden, P., 2016, Sulfur ing worked in Cu-Au and Cu-Mo porphyry depos-
isotope and trace element systematics of zoned pyrite crystals from the El
its, and more extensively in iron oxide copper-gold
Indio Au-Cu-Ag deposit, Chile: Contributions to Mineralogy and Petrology,
v. 171, article 33. deposits, her research focuses on understanding
Tardani, D., Reich, M., Deditius, A.P., Chryssoulis, S., Sánchez-Alfaro, P., the genesis and evolution of hydrothermal deposits from a multidisciplinary
Wrage, J., and Roberts, M.P., 2017, Copper-arsenic decoupling in an active approach integrating petrological, structural, and geochemical work. Cur-
geothermal system: A link between pyrite and fluid composition: Geochi- rently she works as a postdoctoral researcher for the Millennium Nucleus
mica et Cosmochimica Acta, v. 204, p. 179–204. Center of Metals Tracing along Subduction at the University of Chile.

View publication stats