BULLETIN OF THE SOCIETY OF ECONOMIC GEOLOGISTS

Vol. 111 January–February No. 1

The Timing of Magmatism and Ore Formation in the El Abra Porphyry

Copper Deposit, Northern Chile: Implications for Long-Lived Multiple-Event
Magmatic-Hydrothermal Porphyry Systems*
Karen J. Correa,1,† Osvaldo M. Rabbia,2 Laura B. Hernández,2 David Selby,3 and Marcelo Astengo4
1 Departamento de Ciencias de La Tierra, Casilla 160-C, Universidad de Concepción, Concepción CP 4030000, Chile
2 Instituto GEA, Casilla 160-C, Universidad de Concepción, Concepción CP 4030000, Chile
3 Department of Earth Sciences, University of Durham, Durham DH1 3LE, UK
4 Superintendencia de Geología, Sociedad Contractual Minera El Abra, Calama CP 1390000, Chile

Abstract
Zircon LA-ICP-MS and CA-TIMS and molybdenite Re-Os geochronology for the El Abra porphyry cop-
per deposit (Chile) document a ~8.6 Ma protracted magmatic-hydrothermal history. Initial magmatic events
occurred at ~45 Ma, and continued to ~36 Ma. An ~1.8 m.y. magmatic lull is recorded in the early stages of
magmatism, which is characterized by equigranular rocks from the Pajonal suite. Magmatism from ~41 to
~36 Ma, corresponding to the El Abra Granodiorite Complex, is more recurrent, ending in a ~1.4 m.y. period of
porphyritic intrusions, which are coeval with Cu-Mo mineralization. Late porphyritic units reveal subtle zircon
inheritance likely associated with early magmatic pulses. The young stages of magmatism indicate system reju-
venation based on plagioclase phenocryst compositions, suggesting that the porphyry system is ultimately linked
to a less evolved magma. Although there are multiple porphyry and veining events, molybdenite Re-Os ages
deine a focused mineralization episode from 36.34 to 36.18 Ma. Molybdenite Re-Os ages from nearby related
prospects belonging to the El Abra cluster expand the mineralization period in the district up to ~1.0 m.y., sug-
gesting a multistage, long-lived hydrothermal system. Integration of the molybdenite Re-Os dates and those
from previous works at the El Abra and Toki clusters, accompanied by available tectonic reconstruction along
the Domeyko fault system suggest the development, by late Eocene times, of a single, large (~30 km across)
porphyry copper cluster, which was subsequently separated by an offset of ~35 km along the West fault.

Introduction
The El Abra porphyry copper deposit (~8.5 Mt Cu; Camus,
2003) is located in the Central Andes region of northern Chile
(Fig. 1). This ore deposit is part of the middle Eocene to early
Oligocene porphyry copper belt, which spans over 1,400 km
from 18° to 31°S (Camus, 2003). This mineralized belt was
structurally controlled by the transpressional Domeyko fault
system, which shows important along-strike displacements
and shortening (Coira et al., 1982; Mpodozis et al., 1993). The
development of this belt is associated with the cessation of a
compressive tectonomagmatic cycle, when volcanism waned
and porphyry copper systems formed above large silicic
plutonic complexes (Dilles et al., 1997; Ballard et al., 2002;
Haschke, 2002). World-class deposits such as Chuquica-
mata (35–31 Ma), La Escondida (38–37 Ma), and Collahuasi
† Corresponding author: e-mail, rabbia@udec.cl
*A digital supplement is available at http://economicgeology.org/ and at http://
econgeol.geoscienceworld.org/. Appendices 1, 2, and 3 follow the text.
©2016 by Economic Geology, Vol. 111, pp. 1–28
0361-0128/16/4367/1-28
(34–31 Ma), among others, were formed during this metal-
logenic cycle (Camus, 2003).
The El Abra porphyry copper deposit is associated with
the El Abra granodiorite complex (Ambrus, 1977, 1979). The
northern extent of this complex is defined by a sharp contact
with the older Pajonal diorite (Dilles et al., 1997). Together,
these form the Pajonal-El Abra igneous suite. This suite crops
out 2 km eastward of the West fault and ~35 km north of the
Chuquicamata porphyry system (Fig. 1). The Pajonal-El Abra
suite has evolved over an ~7.5 m.y. period, from the oldest
Pajonal diorite (43.2 Ma) to the youngest Lagarto porphyry
(35.6 Ma) (Valente et al., 2006).
Surface geologic mapping within the El Abra deposit has
documented up to seven texturally distinct intrusive units (Bar-
rett, 2004). Crosscutting relationships, along with K-Ar (biotite)
and U-Pb (zircon) ages for some of these units, suggest more
than one period of magmatic-hydrothermal activity (Maksaev
et al., 1990; Dilles et al., 1997; Tomlinson et al., 2001; Bal-
lard et al., 2002; Campbell et al., 2006). Current geochronol-
ogy indicates that magmatic activity within the deposit spans
Submitted: April 22, 2015
1 Accepted: September 6, 2015
 2
Iquique Quebrada Blanca
Collahuasi-Ujina
DFS

Cluster Toki El Abra

22° S Radomiro Tomic
Chuquicamata
Calama
Telégrafo Salar de

Pacific Ocean
Atacama basin

Argenna
Antofagasta
EA-1284 24° S
Zaldívar
La Escondida
EA-657
EA-6
Taltal
EA-916 26° S El Salvador

Potrerillos
EA-1116 Copiapo

EA-903 EXPLANATION
EA-1141 Dark diorite

CORREA ET AL.
EA-710 Central diorite
Equis quartz-monzodiorite
EA-561
Rojo Grande granite
Clara granodiorite
Apolo leucogranite and aplite
El Abra porphyry
Intrusive breccias
Hydrothermal breccias
Colluvium and gravels
El Abra pit
Potassic zone
Quartz-sericite zone
Lineaments
U-Pb Ages
Re-Os Ages

Fig. 1. Surface geology map of the El Abra mine, with sample locations. Projection: UTM19; Datum: Provisional South America 1956 (PSAD56). DFS = Domeyko fault
system. Modiied from an unpublished El Abra mine map.
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
over an ~3.5 m.y. period, between the older equigranular Cen-
tral diorite (40.60 ± 0.30 Ma; U-Pb zircon LA-ICP-MS) and
one dacitic variety of the younger El Abra porphyry (37.16 ±
0.38 Ma; U-Pb zircon LA-ICP-MS) (Ballard et al., 2002; Camp-
bell et al., 2006). Subsequently, a deep (1 km below surface)
drilling program revealed several new texturally distinct por-
phyritic units (Ardila, 2009). Discovery of these new porphyries
provides evidence for multiple intrusive events but also raised a
question about the age of these units. Samples from these deep
drill cores have been dated in this work.
The mineralization age at El Abra has been estimated from
available K-Ar data (Ambrus, 1977; recalculated by Maksaev,
1990; Dilles et al., 1997; Tomlinson et al., 2001) obtained on
hydrothermal biotite and sericite associated with potassic and
quartz-sericite alterations, respectively. These data suggest a
duration of ~2 m.y. (35.0–36.8 Ma) for the hydrothermal system
and its cooling below the K-Ar closure temperature of micas.
Despite the evidence provided by previous field work and
geochronological studies, the duration of the magmatic system
and the precise timing of mineralization remain poorly con-
strained, which limits the understanding of ore formation in
this porphyry copper deposit and in the metallogenic district.
Here we present and discuss the precise timing of ore deposi-
tion and duration of the magmatic-hydrothermal system at the
El Abra porphyry copper deposit using new U-Pb ages obtained
by zircon LA-ICP-MS and CA-TIMS 206Pb/238U analyses (on
the same samples) from the El Abra granodiorite complex,
and high precision Re-Os ages on molybdenite. The integra-
tion of results from previous workers with those presented here
extends the duration of magmatic activity to ~8.6 m.y., conirm-
ing the long-lived nature of the magmatic-hydrothermal system
associated with this deposit and providing better constraints on
the timing of ore formation. The mineralization and emplace-
ment age of porphyritic intrusions can now be correlated with
those from nearby prospects and other systems along the trace
of the Domeyko fault system. We propose the existence, by late
Eocene times, of a large cluster of porphyry copper systems,
which were subsequently bisected and displaced ~35 km apart
along the West fault during post-Eocene times.
The El Abra Granodiorite Complex
The work of many authors over decades and at different scales
has led to some complexity in the nomenclature of rocks of
the El Abra granodiorite complex. For clarity, this will be
briely addressed and the adopted changes in the nomen-
clature for this contribution will be indicated. The El Abra
granodiorite complex can be considered as part of a larger
unit called the Pajonal-El Abra complex. The late Eocene El
Abra Granodiorite Complex is a regional unit composed by a
set of calc-alkaline intrusions of intermediate to felsic compo-
sition (Ambrus, 1977). Subsequently Ambrus (1979) applied
the term El Abra granodiorite complex to a group of granodi-
orite intrusions, and this term has been maintained in subse-
quent studies (e.g., Dilles et al., 1997; Tomlinson et al., 2001;
among others). The El Abra granodiorite complex is largely
emplaced along its south and east sides into lower Mesozoic
metavolcanic and metasedimentary rocks, along its northeast
side into Upper Cretaceous-lower Tertiary sedimentary rocks
of the Tolar Formation, and along its northern side into the
middle Eocene Pajonal monzodioritic complex (Tomlinson
3
et al., 2001). The El Abra granodiorite complex is regionally
divided into Cerro Panizo de Ojuno quartz-diorite, El Abra
granodiorite, and Llareta granodiorite-granite (Tomlinson et
al., 2001). At the El Abra deposit, however, the so-called El
Abra granodiorites are known to be represented by the Cen-
tral diorite, the Dark diorite and the Equis quartz-monzodi-
orite (Barrett, 2004). This nomenclature will be used in this
work. The El Abra granodiorites along with Clara granodio-
rite are the host rocks of the copper mineralization. In the
northeast part of the open pit mine the geology comprises the
Rojo Grande granite, equivalent to the distal Llareta gran-
ite (Tomlinson et al., 2001), which in turn has a transitional
contact with the Llareta granodiorite, both located north of
the open pit mine. All above units are intruded by the Clara
granodiorite, the largest intrusion of this complex, which
crops out in the southern sector of the deposit and is the main
host rock of mineralization at depth. The Clara granodiorite
has been previously divided into two units, Clara and South
granodiorite (Ambrus, 1979); however, subsequent studies
have found no evidence for a contact between these units
and instead group them under the name of Cotari granodio-
rite (Dilles et al., 1997). More recently, the terms “Clara” or
“South” have been used interchangeably to describe the Cotari
granodiorite (e.g., Ballard, 2001; Tomlinson et al., 2001; Bal-
lard et al., 2002; Campbell et al., 2006). The term Clara grano-
diorite is adopted here for this unit. All units mentioned so
far are represented by equigranular rocks, although the Clara
granodiorite may locally be porphyritic. In contrast, the young-
est units are dominantly porphyritic. Three porphyritic intru-
sions occur in the El Abra area: the aplitic granite porphyry,
the main Lagarto granodiorite porphyry, and the late Lagarto
maic granodiorite porphyry (Dilles et al., 1997). The oldest of
these units is represented by a group of dikes and sheets cross-
cutting the Clara granodiorite that includes the aplitic granite
porphyry, the Apolo leucogranite, and the aplite-pegmatite
unit. All these porphyritic intrusions are interpreted to rep-
resent water-saturated, vapor-bubble–bearing, granitic melts
that were ilter-pressed from the upper portion of the adja-
cent, nearly crystalline Clara granodiorite (Tomlinson et al.,
2001). Leucogranites older than the Clara granodiorite have
also been reported (e.g., Ballard et al., 2002). The aplite-peg-
matite unit is mostly present toward the southwest, outside the
open pit mine. The main Lagarto granodiorite porphyry and
the late Lagarto maic granodiorite porphyry (terms used by
Dilles et al., 1997, and Tomlinson et al., 2001) extend from the
El Abra deposit to the Los Volcanes (also known as Conchi or
Conchi Viejo) porphyry Cu-Mo prospect, located ~10 km to
the southeast. Campbell et al. (2006) and Ballard et al. (2001)
used the term El Abra porphyry for a quartz-monzonitic por-
phyry at the mine site equivalent to the Lagarto porphyries of
Dilles et al. (1997). The main Lagarto granodiorite porphyry
is associated with potassic alteration and the late Lagarto
maic granodiorite porphyry postdates the potassic alteration
(Dilles et al., 1997). Tomlinson et al. (2001) suggest that each
type of the two porphyry varieties actually consists of several
porphyries emplaced at slightly different times. Additional
textural varieties of these porphyritic units were intercepted
by deep drill holes (Ardila, 2009). Table 1 summarizes the
nomenclature and temporal relationships between the dif-
ferent rock units described here. The spatial association of
4 CORREA ET AL.

Table 1. Summary of Nomenclature and U-Pb Ages for Rock Units in the El Abra and Chuquicamata Districts

El Abra district Chuquicamata district

Pajonal-El Abra Complex Los Picos-Fortuna Complex

District nomenclature Mine nomenclature Los Picos-Fortuna Complex

Unit names Unit names Age (Ma) Unit names Age (Ma)

Pajonal diorite 42.9 ± 0.51 Los Picos monzodiorite 43.1 ± 0.61

43.21 ± 0.212 42.3 ± 0.21
41.84 ± 0.273
Cerro Panizo de Ojuno quartz Dark diorite 40.2 ± 0.51
diorite 40.83 ± 0.342
40.51 ± 0.993
Abra granodiorite Central diorite 40.6 ± 0.31
40.64 ± 0.252
40.62 ± 0.313
Equis quartz-monzodiorite 41.74 ± 0.192
38.5 ± 0.41
Llareta granodiorite and granite Rojo grande granite 39.14 ± 0.242 Antenna granodiorite 38.82 ± 0.643
39.09 ± 0.034 Fortuna Gris 39.1 ± 0.41
Clara granodiorite Clara granodiorite 38.71 ± 0.114 Fiesta granodiorite 39.38 ± 0.483
38.0 ± 0.31 Fortuna Clara 38.5 ± 0.41
37.55 ± 0.202
37.93 ± 0.423
Apolo aplites, leucogranite, Apolo aplites, leucogranite, 37.55 ± 0.262 Tetera porphyries
aplitic porphyries aplitic porphyries 38.5 ± 0.41
San Lorenzo porphyry 38.0 ± 0.41
38.23 ± 0.801
Lagarto granodiorite porphyry Dacite 37.49 ± 0.104 Opache porphyry 37.3 ± 11
porphyry 37.4 ± 0.31 37.9 ± 0.21
El Abra 36.89 ± 0.312 37.67 ± 0.623
poprhyry 37.16 ± 0.383
Lagarto maic granodiorite Rhyolite 36.07 ± 0.154
porphyry porphyry

1 Ballard (2001), LA-ICP-MS

2 Valente (2008) in Santana (2010), LA-ICP-MS
3 Campbell et al. (2006), LA-ICP-MS
4 This study, CA-TIMS

significant copper mineralization with the young porphyritic
units suggests a genetic link between ore formation and the
final stages of the magmatic evolution of this large igneous
complex (Ambrus, 1977; Dilles et al., 1997; Tomlinson et al.,
2001; Ballard et al., 2002; Barrett, 2004).
Hydrothermal Alteration and Mineralization at
the El Abra Deposit
The main alteration zones at El Abra include a potassic core,
locally more albitic at depth, surrounded by a quartz-sericite
zone with an outer envelope of poorly developed propylitic
alteration (Fig. 1). Hydrothermal alteration is temporally
and spatially associated with intrusion of the El Abra por-
phyries (main Lagarto and maic Lagarto), dikes, and igne-
ous-hydrothermal breccias (Tomlinson et al., 2001; Barrett,
2004; Ardila, 2009). Potassic alteration highlights areas with
pervasive replacement of plagioclase by K-feldspar, recrys-
tallization of biotite phenocrysts, and presence of “shreddy”
biotite intergrown with magnetite and relict titanite. Miner-
alization associated with potassic alteration is zoned. In the
central and deeper parts of the mine, bornite + chalcopyrite
± magnetite associations prevail, followed by chalcopyrite >
bornite ± molybdenite, with chalcopyrite > pyrite ± molyb-
denite dominant in the outer portions of the mineralization
(Ardila, 2009). Albitic alteration is localized in the west and
southwest sectors of the open pit where it occurs in two main
styles: (1) small breccias and dikes associated with tourmaline
in thin veinlets with an albitic halo, or (2) in a breccia matrix
in the deepest parts of El Abra deposit. Sulides are rare in
albitic alteration, but when present they include pyrite and
chalcopyrite. Quartz-sericite-clay alteration overprints the
potassic and albitic zones and occurs in hydrothermal brec-
cias and zones associated with northwest-trending structures
and to the southeast along the main structural trend in the
district. These structures mainly correspond to D-veins with
variable thickness (1 mm–1 m) and with sericite, chlorite, and
quartz halos. Quartz-chalcopyrite ± bornite ± pyrite ± molyb-
denite, locally along with tourmaline and anhydrite, are the
typical vein-illing minerals. A proximal propylitic alteration,
characterized by chlorite, epidote, and actinolite, is developed
near the periphery of the deposit. K-Ar dating of biotite from
potassic alteration yields ages ranging from 35.0 to 35.3 Ma
(Ambrus, 1977; Maksaev, 1990; Tomlinson et al., 2001), with
K-Ar determinations on sericite and sericitized whole-rock
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
providing similar ages of 36.0 ± 1.3, 35.7 ± 3.3, and 34.5 ±
1.4 Ma (Tomlinson et al., 2001).
The NW-SE to WNW-ESE structures control the emplace-
ment of the mineralization at El Abra, including porphyry
dikes and (A, AB) veinlets. D-type veinlets, related to quartz-
sericite alteration show an extensional behavior, while the
postmineral structures (trending NW-SE to WNW-ESE,
NNW-SSE and E-W at the mine site), match the axes of prin-
cipal stresses controlling the sinistral movement of the West
fault (Santana, 2010). Also, postmineral structures are reacti-
vations of D-type veinlets.
Sampling and Analytical Methodology
Petrography and plagioclase mineral chemistry and
microtextures
This study examined more than 9,000 m of diamond drill core
from the deep (800–2,000 m) levels of the El Abra deposit. In
addition to petrography we also document the microtextures
of plagioclase using the electron microprobe (JEOL JXA8600
Superprobe provided with a Bruker SDD EDS spectrometer,
model XFLASH 5010, with 123 eV resolution at Mn Ka) at
Instituto de Geología Económica Aplicada, Concepción Uni-
versity, Chile. Two varieties of porphyritic samples were ana-
lyzed, including a rhyolite porphyry (EA-1284, also dated by
U-Pb) and a dacite porphyry (EA-874). Selected phenocrysts
were analyzed along a core-rim transect choosing the most
representative domains for each grain using high-contrast,
backscattered electron (BSE) imaging. The analytical condi-
tions can be summarized as follows: major elements were mea-
sured using 15 kV, 20 nA, 10-µm beam diameter, and peak/
A
Dac. P Rhy. P
Dac. P
Rojo Grande
granite
0 6cm
Rhy. P
Fig. 2. A. Inclusions of dacite porphyry (Dac. P) in rhyolite porphyry (Rhy. P). B. Multiple phases of B- and D-veinlets cross-
cutting the rhyolite porphyry and Clara granodiorite (Clara grd.).
5
background counting time 20/5 s. Trace elements (Sr, Fe, K,
and Mg) were measured on the same spot after major ele-
ment measurement using peak/background counting time up
to 400/200 s (depending on the element concentration) and a
beam current of 40 nA. Detection limits are 184 ppm for Sr,
450 ppm for Ba, 295 ppm for Fe, and 70 ppm for Mg. Ba and
Mg were below detection limits in most analyses, so they were
no longer considered. Calibration was based on the following
standards: anorthite USNM 137041 (Al), fayalite USNM 85276
(Fe), microcline USNM 143966 (Si and K) (Jarosewich et al.,
1980), forsterite (Mg), barite (Ba), and wollastonite (Ca). Data
reduction was performed using the ZAF correction method. A
summary of results is displayed in Table 2.
U-Pb zircon dating
High-precision dates, i.e., CA-TIMS 206Pb/238U zircon ages,
are not available for the El Abra district. Aware about the dis-
crepancy among different U-Pb methods (e.g., LA-ICP-MS
and CA-TIMS; von Quadt et al., 2011; Chiaradia et al., 2013,
2014) and taking into account that the bulk of the geochrono-
logical database compiled for the El Abra district corresponds
to LA-ICP-MS ages (Table 1), we decided to replicate ages
using both U-Pb methods (on the same sample).
Two representative samples from the El Abra porphyry
were selected for U-Pb zircon dating, including a dacite (EA-
561) and a rhyolite (EA-1284). Both of these samples were
selected from deep drill core (500–1,300 m). The dacite
porphyry sample came from a small body; locally, small
(3–15 cm) inclusions of this porphyry are present as enclaves
in the rhyolite porphyry, a relationship observed at different
sites in the mine (Fig. 2A). Both porphyries intrude the Clara
Dac. P
Dac. P
Dac. P
Rhy. P
0 6cm
Clara grd.
 6
Table 2. Summary of Microtextures in Plagioclase Phenocrysts of Dacitic and Rhyolitic Porphyries (EA-874 and EA-1284)

Texture/unit Interpretation BSE image BSE image Compositional proile

Sieve-texture (ST)/ Partial dissolution due to

dacite porphyry reaction with more An (%) Sr and Fe (ppm)
Ca-rich melt
Core Rim
3000
50
ST 2000
30 1000
10 0
500 µm 125 µm 0 300 600 900 1200
Resorption surfaces (RS)/ Dissolution while reacting
dacite porphyry- with a more primitive
rhyolite porphyry magma
Core Rim
60 2000
RS
1500
40
1000
20 500
0 0

CORREA ET AL.
Rim 0 200 400
400 µm 100 µm
Core dissolution and patchy Dissolution while reacting An (%) Sr and Fe (ppm)
texture (PT) / dacite with a more primitive Core Rim
porphyry magma RS 3000
50
2000
PT 30 1000

10 0
0 50 100 150 200
100 µm 100 µm
approx. distance from the core µm
Fine-scale oscillatory zoning Convection driven small
(FOZ)/dacite porphyry- scale physical-chemical
rhyolite porphyry perturbation at the crystal-
melt interface
FOZ

100 µm 30 µm

Interpretation of microtextures based on Renjith (2013) and Pietranik et al. (2006)

Yellow squares show the extended area; blue, orange, and green solid lines represent An%, Fe (ppm), and Sr (ppm), respectively; green and orange dotted lines are minimum detection limits for Sr
(184 and Fe 295 ppm), respectively.
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE 7

granodiorite. In addition, three equigranular samples from
older units were selected, including the Rojo Grande granite
(EA-657), Clara granodiorite (EA-1141), and an amphibole-
rich variety of the Clara granodiorite (EA-1116). Crosscutting
relationships between the Rojo Grande granite and the Clara
granodiorite are marked by Schlieren and aplitic zones sug-
gesting that the Rojo Grande granite was not entirely crys-
talline when the Clara granodiorite intruded (Tomlinson et
al., 2001). Contacts between the Clara granodiorite and the
amphibole-rich variety are ambiguous and cannot be easily
determined in drill core. Sample locations are shown in Fig-
ure 1 and their coordinates are listed in Table 3.
About 25% of analyzed zircons were screened by cathodo-
luminescence (CL). CL images show textural characteristics
indicating internal growth zonation (App. 1). Occasionally CL
images reveal the presence of older cores (App. 1) and min-
eral inclusions identiied as apatite (App. 2).
U-Pb zircon dating on selected samples was performed
using LA-ICP-MS and CA-TIMS methodologies at the Paciic
Centre for Isotopic and Geochemical Research (PCIGR),
University of British Columbia, Canada. The full U-Pb data-
set is provided in an Excel table online as a digital supplement
(digital supp.).
LA-ICP-MS
The LA-ICP-MS zircon analyses were carried out follow-
ing the methodology described by Tafti et al. (2009). Instru-
mentation employed for LA-ICP-MS dating of zircons at the
PCIGR comprises a New Wave UP-213 laser ablation sys-
tem and a Thermo Finnigan™ Element 2™ single collector,
double-focusing, magnetic sector ICP-MS. All zircons greater
than about 50 µm in diameter were picked from the mineral
separates and were mounted in an epoxy puck along with sev-
eral grains of the 337.13 ± 0.13 Ma Plešovice zircon standard
(Sláma et al., 2007), together with a Temora 2 reference zir-
cons (416.78 ± 0.33 Ma), and brought to a very high polish.
The surface of the mount was washed for 10 min with dilute
nitric acid and rinsed in ultraclean water prior to analysis. The
highest quality portions of each grain, free of alteration, inclu-
sions, or possible inherited cores, were selected for analysis.
Line scans rather than spot analyses were employed in order
to minimize elemental fractionation during the analyses. A
laser power level of 38% and a 30 µm spot size were used.
Backgrounds were measured with the laser shutter closed
for 10 s, followed by data collection with the laser iring for
approximately 35 s. The time-integrated signals were analyzed
using GLITTER software (van Achterbergh et al., 2001; Grif-
in et al., 2008), which automatically subtracts background
measurements, propagates all analytical errors, and calculates
isotopic ratios and ages. Corrections for mass and elemental
fractionation were made by bracketing analyses of unknown
grains with replicate analyses of the Plešovice zircon standard.
A typical analytical session at the PCIGR consists of four anal-
yses of the Plešovice zircon standard, followed by two analyses
of the Temora 2 zircon standards, ive analyses of unknown
zircons, two standard analyses, ive unknown analyses, etc.,
and inally two Temora 2 zircon standards and four Plešovice
standard analyses. Temora 2 reference zircon was analyzed as
an unknown in order to monitor the reproducibility of the age
determinations on a run-to-run basis (see digital supp.). Final
interpretation and plotting of the analytical results employed
the ISOPLOT software of Ludwig (2003).

Table 3. Summary of LA-ICP-MS (weighted mean age), CA-TIMS (youngest grain age), and Coordinates for Each Dated Sample

UTM coordinates1 LA-ICP-MS CA-TIMS

Unit Sample North East Elevation 206Pb/238U Ma 206Pb/238U Ma n

Rojo Grande granite EA-657 7576599.78 516599.79 3965.6 38.75 ± 0.61 39.09 ± 0.03 5
MSWD = 0.15 MSWD = 1.4
n = 20, 4 of 20 rej. n=4
102.07 ± 0.151
Clara granodiorite EA-1141 7576199.97 516900.15 3804.93 37.82 ± 0.57 38.71 ± 0.11 4
MSWD = 0.30
n = 19, 1 of 19 rej.
Mesocratic Clara granodiorite EA-1116 7576300.22 517100.22 3805.22 38.83 ± 0.47 37.45 ± 0.09 5
MSWD = 0.39
n = 19, 1 of 19 rej.
El Abra dacite porphyry EA-561 7576000.15 517399.44 3804.64 38.94 ± 0.65 37.49 ± 0.10 5
MSWD = 0.59
n = 18, 3 of 18 rej.
177.6 ± 3.72
El Abra rhyolite porphyry EA-1284 7576748.52 517848.51 3774.39 38.48 ± 0.62 36.07 ± 0.15 5
MSWD = 1.7
n = 19, 2 of 19 rej.
236.91 ± 2.12 202.35 ± 0.301
126.95 ± 0.271

Data for old xenocrystic core ages are included for each method; coordinates: projection-UTM 19K; Provisional South American Datum 1956; rej = rejected
1 Collar coordinates
2 Old zircon xenocrysts
8 CORREA ET AL.
Chemical abrasion-thermal ion mass spectrometry
(CA-TIMS)
CA-TIMS procedures described here are modiied from
Mundil et al. (2004); Mattinson (2005), and Scoates and
Friedman (2008). After rock samples have undergone stan-
dard mineral separation procedures, zircons are handpicked
in alcohol. The clearest, crack- and inclusion-free grains are
selected, photographed and then annealed in quartz glass
crucibles at 900°C for 60 h. Annealed grains are transferred
into 3.5 mL PFA screwtop beakers, ultrapure HF (up to
50% strength, 500 µL) and HNO3 (up to 14 N, 50 µL) are
added and caps are hand tightened. The beakers are placed
in 125 mL PTFE liners (up to four per liner) and about 2 mL
HF and 0.2 mL HNO3 of the same strength as acid within
beakers containing samples are added to the liners. The liners
are then slid into stainless steel Parr™ high pressure dissolu-
tion devices, which are sealed and brought up to a maximum
of 200°C for 8 to 16 h (typically, 175°C for 12 h). Beakers
are removed from liners and zircon is separated from leach-
ate. Zircons are rinsed with >18 MΩ.cm water and subboiled
acetone. Then 2 mL of subboiled 6N HCl is added and bea-
kers are set on a hotplate at 80° to 130°C for 30 min and again
rinsed with water and acetone. Masses are estimated from
the dimensions (volumes) of grains. Single grains are trans-
ferred into clean 300 µL PFA microcapsules (crucibles), and
50 µL 50% HF and 5 µL 14 N HNO3 are added. Each zir-
con is spiked with a 233-235U-205Pb tracer solution (EARTH-
TIME ET535), capped and again placed in a Parr liner (8–15
microcapsules per liner). HF and nitric acids in a 10:1 ratio,
respectively, are added to the liner, which is then placed in
the Parr high pressure device and dissolution is achieved at
240°C after 40 h. The resulting solutions are dried on a hot-
plate at 130°C, 50 µL 6N HCl is added to microcapsules and
luorides are dissolved in high pressure Parr devices for 12 h
at 210°C. HCl solutions are transferred into clean 7 µL PFA
beakers and dried with 2 µL of 0.5 N H3PO4. Samples are
loaded onto degassed, zone-reined Re ilaments in 2 µL of
silicic acid emitter (Gerstenberger and Haase, 1997). Isotopic
ratios are measured in a modiied single collector VG-54R or
354S (with Sector 54 electronics) TIMS equipped with ana-
logue Daly photomultipliers. Analytical blanks are 0.2 pg for
U and 1.0 pg for Pb. Uranium fractionation was determined
directly on individual runs using the EARTHTIME ET535
mixed 233-235U-205Pb isotope tracer and Pb isotope ratios were
corrected by fractionation by 0.25%/amu, based on replicate
analyses of the NBS-982 reference material and the values
recommended by Thirlwall (2000). Data reduction employed
Table 4. Sample Coordinates and Results of Re-Os Molybdenite Dating (2σ level)
UTM Coordinates
Sample Vein/ unit North East wt (g) Re (ppm)
EA-916 D/EADPP 7576460 517959 0.010 508.2
EA-903 B/EADP 7576200 517400 0.012 242.8
EA-710 B/EARP 7576100 516700 0.012 300.7
Coordinates: Projection-UTM 19K; Provisional South American Datum 1956
EADP = El Abra dacite porphyry; EARP = El Abra rhyolite porphyry
1 Includes decay constant
the Excel-based program of Schmitz and Schoene (2007).
Standard concordia diagrams were constructed and regres-
sion intercepts, as well as weighted averages were calculated
with Isoplot (Ludwig, 2003). Unless otherwise noted all errors
are quoted at the 2σ or 95% level of conidence. Isotopic dates
are calculated with the decay constants λ238U = 1.55125E-10
and λ235U = 9.8485E–10 (Jaffey et al., 1971). Three EARTH-
TIME U-Pb (ET100 Ma) synthetic solutions were analyzed
to monitor the accuracy of CA-TIMS results. The averages of
those three runs are as follows:
206Pb/238U± random (+tracer) [+λ]
= 100.144 ± 0.07 (0.12) [0.16] Ma; MSWD = 0.06
207Pb/235U ± random (+tracer) [+λ]
= 100.19 ± 0.16 (0.19) [0.23] Ma; MSWD = 0.24
207Pb/206Pb ± random [+λ]
= 100.4 ± 3.0 [3.99] Ma; MSWD = 0.23
Re-Os in molybdenite
Samples for Re-Os dating in molybdenite were chosen tak-
ing into account the type of veinlet and their distribution
within the deposit (following a SW-NE transect, see Fig. 1;
for coordinates, see Table 4). Analyzed molybdenites come
from B- and D-type veinlets hosted by El Abra porphyry
units. Sample EA-916 (dacite porphyry) corresponds to a 3- to
7-mm-thick D-veinlet, mainly illed by quartz with molybde-
nite bands toward the edges. It presents an up to 6-mm-wide
quartz-sericite alteration halos, which locally follows previous
thin AB-veinlets with chalcopyrite and quartz-potassium feld-
spar ine halos. Grain size of molybdenite varies from <1 to 4
mm. Samples EA-903 (dacite porphyry) and EA-710 (rhyolite
porphyry) correspond to a 5- to 10-mm-thick B-type vein-
lets, consisting of quartz and molybdenite (grain size <1–3
mm) present as aggregates and/or bands preferentially at
the edge of the veinlet, with disseminated chalcopyrite (Fig.
2B). Analyses were performed in the Laboratory for Sulide
and Source Rock Geochronology and Geochemistry in the
Durham Geochemistry Center at the University of Durham
(UK). The complete analytical procedure for Re-Os determi-
nations is described in Selby and Creaser (2004) and Selby et
al. (2007), and it is only briely summarized here. Molybde-
nite samples were dissolved and equilibrated with a known
amount of 185Re and isotopically normal Os in inverse aqua
regia (2:1 16 N HNO3 and 12 N HCl, 3 mL) at 240°C for 24 h
in a Carius tube. Rhenium and Os were isolated and puriied
by solvent extraction, microdistillation, and anion exchange
± 187Re (ppm) ± 187Os (ppb) ± Age (Ma) ± ±1
2.5 319.4 1.6 193.4 0.9 36.34 0.15 0.18
1.1 152.6 0.7 92.0 0.4 36.18 0.15 0.18
1.4 189.0 0.9 114.1 0.5 36.22 0.15 0.18
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE 9

chromatography, and analyzed by negative TIMS on a Fisher
Scientiic TRITON mass spectrometer using Faraday col-
lectors. Total procedural blanks for Re and Os, are 2 pg and
0.5 pg, respectively, with an 187Os/188Os blank composition
of 0.24 ± 0.02 (n = 6). Rhenium and Os concentrations and
Re-Os molybdenite date uncertainties are presented at the
2σ level, which includes the uncertainties in Re and Os mass
spectrometer measurement, spike and standard Re and Os
isotope compositions, and calibration uncertainties of 185Re
and 187Os. Because a mixed 185Re and Os tracer solution is
used, uncertainties in weights of sample and tracer solution
do not affect the calculated age, and are not considered. How-
ever, sample and tracer solution weight uncertainties are con-
sidered in determining the uncertainty in the Re and 187Os
concentrations. Uncertainty with and without the 187Re decay
constant (Smoliar et al., 1996; Selby et al., 2007) is also con-
sidered (Table 4).
Results and Discussion: Magmatism
The discrepancy among the CA-TIMS and LA-ICP-MS
results (Fig. 3A, B) is discussed below case by case. We can
anticipate, however, that in three cases (samples EA-1116,
EA-561, and EA-1284), there is clearly a demonstration of
the fact that LA-ICP-MS ages are “artiicially” systematically
older than the youngest CA-TIMS age as shown already by
von Quadt et al. (2011) and Chiaradia et al. (2013, 2014).
Rojo Grande granite (EA-657)
The oldest analyzed rock corresponds to the Rojo Grande equi-
granular granite. This is a leucocratic, phaneritic rock (Fig. 4A,

A. CA-TIMS B. LA-ICP-MS

Fig. 3. Ages of intrusive rocks from the El Abra deposit. A. CA-TIMS 206Pb/238U dates. Best estimated ages (youngest grain),
or mean weighted age for equivalent zircon population, are indicated for each unit. B. LA-ICP-MS 206Pb/238U dates with
probability plots and density functions. Light gray histograms are from the density plot and dark gray columns are 206Pb/238U
mean weighted ages. Rej. = rejected data points (black bars).
10 CORREA ET AL.

A B

1cm
0 3cm

C D

1cm
0 3cm

E F

1cm
0 3cm

G H

1cm
0 4cm

I J

1cm
0 3cm

Quartz Amphibole(Hornblende)
Feldspars Magnetite
K-Feldspars Titanite
Biotite/Phlogopite Ilmenite
Fig. 4. False color QEMSCAN and hand-sample photos of dated samples. A, B. EA-657, Rojo grande granite. C, D. EA-1141,
Clara granodiorite. E, F. EA-561, dacite porphyry. G, H. EA-1116, mesocratic Clara granodiorite. I, J. EA-1284, rhyolite
porphyry.
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE 11

B). The major constituents are subhedral plagioclase (34.56
vol %), commonly zoned and slightly altered to sericite, anhe-
dral K-feldspar (28.65 vol %) with perthitic texture, and anhe-
dral quartz crystals (27.50 vol %). Additionally, minor biotite
(6 vol %), mostly secondary, is present in association with
chlorite and magnetite. Accessory phases include rhombic
titanite, magnetite (locally replaced by hematite with ilmen-
ite exsolutions), apatite, and zircon. Zircons are euhedral,
typically smaller than 50 µm, and are common inclusions in
plagioclase. This sample has thin D-veinlets of quartz-chlo-
rite-chalcopyrite with quartz-sericite halos up to 3 mm thick.
The cumulative probability plot method (e.g., Campbell et
al., 2006; see also Chiaradia et al., 2014) was used for removal
of spot analyses representing subtle inheritance and Pb-loss, a
common type of open-system behavior in zircons, from each
LA-IPC-MS data population (digital supp.). The LA-ICP-MS
data have a nearly unimodal distribution (Fig. 3B), yielding
a 206Pb/238U weighted mean age of 38.75 ± 0.61 Ma (mean
square of weighted deviates, MSWD = 0.15) with 4 out of 20
data points rejected (black bars, Fig. 3B). The small MSWD
value may suggest the analytical errors are likely to be over-
estimated. On the other hand, if rejected data are included
(n = 20), the weighted mean becomes 39.40 ± 0.63 Ma with
a still acceptable MSWD of 1.2. However, in the cumulative
probability plot, the four removed zircon data form a distinc-
tive, slightly older (~40.5–42.5 Ma) subpopulation (see digital
supp.) and therefore we prefer to reject these data, despite
this decreases the MSWD value below 1. The CA-TIMS
methods (Fig. 5A, B shows concordia diagrams, respectively)
yields a weighted mean 206Pb/238U date of 39.088 ± 0.026 Ma
(MSWD = 1.4; Fig. 6). Therefore, the LA-ICP-MS age of
Rojo Grande granite is overlapping within the large uncer-
tainty of the LA-ICP-MS method (38.75 ± 0.61 or 39.40 ±
0.63 Ma) to the CA-TIMS 206Pb/238U ages. These dates are
in perfect agreement with previously reported ages for this
unit (Table 1). The Rojo Grande granite unit shows evidence
of a subtle inheritance (slightly older subpopulation in the
cumulative probability plot, digital supp.) from proto-plutons

A B

C D

Fig. 5. Concordia diagrams, left side: LA-ICP-MS, right side: CA-TIMS.

12 CORREA ET AL.

E F

G H

I J

Fig. 5. (Cont.)
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
Fig. 6. Weighted mean 206Pb/238U CA-TIMS dates of Rojo Grande granite (EA-657).
or slightly early magmatic products. An old inherited zircon
core from this sample yields an Early Cretaceous age of 102 ±
0.15 Ma (CA-TIMS).
Clara granodiorite (EA-1141)
The Clara granodiorite is a phaneritic, leucocratic, equigran-
ular rock that is macroscopically characterized by irregular
aggregates of biotite (2–3 mm), biotite-replaced hornblende
(euhedral, up to 6 mm) and prismatic plagioclase surrounded
by K-feldspar (Fig. 4C, D). Plagioclase (54.25 vol %) shows
moderate sericitization, patchy texture, oscillatory zoning,
and locally myrmekitic texture with anhedral quartz inclu-
sions. Anhedral quartz (18.02 vol %) crystallized late, illing
interstices. Potassium feldspar (14.76 vol %) is anhedral with
perthitic texture. Biotite (5%) and amphibole (3%) are minor
constituents partially or totally replaced by secondary biotite
and chlorite. Titanite, magnetite, apatite, and zircon are com-
mon accessory minerals. Titanite occurs as subhedral crystals
partially altered to rutile. Magnetite partially oxidized to hema-
tite is associated with biotite and titanite. Apatite appears as
tiny prismatic crystals usually included in plagioclase. Zircon
tends to be euhedral and appears included in plagioclase and
biotite, as well as associated with magnetite aggregates. Crys-
tal sizes vary between 10 and 50 µm. This sample contains
early quartz veinlets (1–8 mm) with a discontinuous suture of
bornite and K-feldspar and biotite toward the vein margins.
Despite the fact that LA-ICP-MS data from this sample have
a unimodal distribution (Fig. 3B), the cumulative probability
plot shows a complex distribution of data (digital supp.). One
out of 19 zircons analyzed was rejected. It was interpreted to
be a slightly older inherited grain (~40.5 Ma) using the cumu-
lative probability plot (digital supp.). The remaining 18 zircons
yield a weighted mean 206Pb/238U LA-ICP-MS age of 37.82 ±
0.57 Ma (MSWD = 0.30). The same applies to the oldest and
13
discordant 206Pb/238U CA-TIMS date (45.54 ± 0.06 Ma) inter-
preted to represent an inherited zircon grain. The remain-
ing three CA-TIMS dates are not statistically equivalent and
range from 38.91 ± 0.06 to 38.71 ± 0.11 Ma (Figs. 3A, 5D)
(weighted mean CA-TIMS age 38.87 ± 0.24  Ma, MSWD =
4.8). The difference between the youngest CA-TIMS date
(38.71 ± 0.11 Ma, Fig. 3A), which is the closest approximation
to the intrusion emplacement age (Chiaradia et al., 2013) and
the LA-ICP-MS weighted mean age (37.82 ± 0.57 Ma) for the
Clara granodiorite may result either from undetected Pb loss
(see the complex cumulative probability plot, digital supp.) in
LA-ICP-MS ages or from undersampling in CA-TIMS analy-
ses, or a combination of both. Our best estimated CA-TIMS
age for the Clara granodiorite intrusion (38.71 ± 0.11 Ma) is
older than previously reported ages by conventional U-Pb and
LA-ICP-MS dating, which range from 37.5 ± 0.5 to 38.0 ±
0.3 Ma (Table 1; Fig. 7; Dilles et al., 1997; Ballard et al., 2001;
Campbell et al., 2006). However, considering that sampling
sites (and depths) were different in each case, and taking into
account the batholithic size of the Clara granodiorite (Dilles
et al., 1997), it is possible that the observed age range in this
unit relects a long-lived magma system, rather than only dif-
ferences in dating methods.
Mesocratic Clara granodiorite (EA-1116)
The amphibole-rich, mesocratic Clara granodiorite (Fig. 4E,
F) represents a variety of coarse-grained equigranular rock
intercepted at depths of over 1 km from the present pit bot-
tom level. It has a particularly distinctive texture visually domi-
nated by large (up to 20 mm) euhedral, subequant hornblende
crystals. It appears as up to 0.5 m wide meso- to melanocratic
bands in the Clara granodiorite without a clearly deined
intrusive contact. Large amphibole crystals (up to 20  mm)
have previously been described in the Clara granodiorite
14 CORREA ET AL.

Age references: Geochemical data

This work U-Pb (Zircon/CA-TIMS youngest grain)
This work Re-Os (Molybdenite)
This work U-Pb (Zircon/LA-ICP-MS)
This work, subtle inherited grains (LA-ICP-MS)
Tomlinson et al., 2001, K-Ar (Biotite)
Rivera and Pardo, 2004, U-Pb (Zircon/SHRIMP)
Proffet, 2008 in Cabrera, 2011, U-Pb (Zircon/SHRIMP)
Perelló et al., 2012, U-Pb (Zircon/LA-ICP-MS)
Rojas, 2015, U-Pb (Zircon/LA-ICP-MS)
references:
Ballard et al., 2001, U-Pb (Zircon/LA-ICP-MS) This work
Ballard et al., 2001, U-Pb (Zircon/SHRIMP) Ballard et al., 2001
Campbell et al., 2006, U-Pb (Zircon/LA-ICP-MS) Cabrera, 2011
Dilles et al., 1997, U-Pb (Zircon/Conventional) Rojas, 2015
Barra et al., 2013, U-Pb (Zircon/LA-ICP-MS)
Barra et al., 2013, Re-Os (Molybdenite)
Valente, 2008 in Santana, 2010, U-Pb (Zircon/LA-ICP-MS)
Perelló et al., 2012, Re-Os (Molybdenite)

Fig. 7. Compilation of dates in a timeline for the El Abra and Chuquicamata districts, by different methods and authors.
Errors in U-Pb ages are reported at the 2σ level. Uncertainties in Re-Os ages are reported with a 2σ uncertainty. D = diorite,
Grd = granodiorite, mzD = monzodiorite, P = porphyry, Gr = granite, Mesoc. = mesocratic, RT = Radomiro Tomic, PFF =
ine porphyry (minor facies) (Campbell et al., 2006), PFE = Este porphyry (major facies) (Campbell et al., 2006).
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
(Dilles et al., 1997), but with a lower modal abundance than
the melanocratic bands described in this study, which have
variable amphibole contents, commonly greater than 10 wt
%. Biotite-altered diorite xenoliths of uncertain origin (Dark
or Central) are locally hosted by these melanocratic bands in
the Clara granodiorite. Sample EA-1116 consists of euhedral
prismatic crystals of amphibole (11.24 vol %), 2 to 16  mm
in size, partially replaced by secondary biotite and chlorite,
subhedral plagioclase (34.65 vol %), often surrounded by
anhedral K-feldspar (13.51 vol %), and quartz (17.93 vol %).
Secondary biotite aggregates (11.45 vol %) predominate over
primary biotite. Magnetite (2 vol %), relict titanite (± rutile),
apatite, and zircon are present as accessory phases. Large
(20–100 µm) euhedral zircons occur as inclusions in horn-
blende, and less frequently in plagioclase microphenocrysts,
associated with quartz-magnetite aggregates. Some zircon
crystals display thin Th-, U-, and Hf-enriched growth bands
(if viewed in cross section) as part of the compositional zon-
ing and small (5–20 µm) Cl-F-OH apatite inclusions are often
present (see App. 2). Also present are quartz-rich A-type and
AB-type veinlets with K-feldspar, abundant anhydrite, born-
ite, and chalcopyrite.
The 206Pb/238U LA-ICP-MS weighted mean age is 38.83 ±
0.47 Ma (MSWD = 0.39). The LA-ICP-MS data have an almost
unimodal distribution, with only one rejected discordant zir-
con (45.70 ± 1.71 Ma) out of 19 (Fig. 5E). The removed zircon
is interpreted as an inherited grain (digital supp.). Concordia
diagrams are displayed in Figure 5E and F. The ive 206Pb/238U
CA-TIMS dates are not statistically equivalent and range from
38.47 ± 0.08 to 37.45 ± 0.09 Ma (weighted mean age is 38.01
± 0.49, MSWD = 101). The LA-ICP-MS weighted mean is out
of the CA-TIMS ages range, even though most CA-TIMS data
are in the range of the LA-ICP-MS data population (Fig. 3A,
B). The youngest CA-TIMS date (37.45 ± 0.09 Ma) is the best
approximation of the age of intrusion. This CA-TIMS age is
among the youngest reported ages for the Clara granodiorite
(Table 1, Fig. 7) and the full range of CA-TIMS ages (~1 Ma)
covers almost the entire range of reported ages for this unit
(Dilles et al., 1997; Ballard et al., 2001; Campbell et al., 2006;
Valente, 2008, in Santana, 2010).
Dacite porphyry (EA-561)
The selected dacite porphyry (sample EA-561) comes from
a dike that intrudes the Clara granodiorite. It is a mesocratic,
porphyritic rock (Fig. 4G, H) with 50 vol % phenocrysts
(1–4  mm) in a ine-grained groundmass (0.08–0.15  mm)
composed of plagioclase, quartz, and K-feldspar. Plagioclase
phenocrysts (48.54 vol %) are subhedral to euhedral with
K-feldspar rims. Plagioclase phenocryst cores have com-
positions that range from An36 to An29, and also contain Sr
(1,140–1,800 ppm) and Fe (1,100–1,700 ppm). Patchy tex-
ture is observed, with and without dissolution zones and with
ine-scale oscillatory zoning. Sieve resorption surfaces are fol-
lowed by zones with variable, but commonly high increments
in anorthite content (An53-An31) and accompanied by incre-
ments of ~100 ppm of Sr and up to ~500 ppm of Fe. Toward
the rims, anorthite content decreases down to An21-An7, with
Sr below 1,200 ppm and Fe below 900 ppm (Table 2). Primary
biotite phenocrysts (1.3 vol %) are small (0.5 mm) and appear
as subhedral to euhedral single crystals, locally replaced by
15
agglomerates of secondary biotite and epidote. Subrounded
quartz eye phenocrysts (<1 vol %) are small (0.5–2 mm) and
some show embayments. Magnetite is a minor constituent
(2 vol %) along with apatite, titanite, and zircon. Zircons are
euhedral to subhedral and frequently included in plagioclase
and hornblende phenocrysts, but are also locally found as part
of the groundmass. Their average size is about 36 µm. Chal-
cocite, bornite, and digenite are disseminated throughout the
groundmass; locally, a bornite-chalcocite “myrmekite”-type
texture is observed. Quartz-rich A- and AB-type veinlets with
biotite-bornite-chalcopyrite-K-feldspar halos are present in
this sample.
The LA-ICP-MS method yields a weighted mean 206Pb/238U
age of 38.94 ± 0.65 Ma (MSWD = 0.59), from 15 analyzed zir-
cons. Three data were rejected using the cumulative probabil-
ity plot (digital supp.), two of these being slightly older (44.70
± 0.53 and 42.70 ± 0.53 Ma) zircons interpreted as inherited
grains. The older rejected zircon age is concordant while the
youngest one is slightly discordant, possibly due to Pb loss
(Fig. 5G). If both data are considered the weighted mean age
becomes 41.1±1.3 with MSWD = 6.2, a value that indicates
scatter well above the expected analytical uncertainty. Addi-
tionally, both data plot outside the weighted mean age range
(black bar, Fig. 3B). The third rejected result is from an old
zircon core that yielded a Jurassic age of 177.6 ± 3.1 Ma (digi-
tal supp.). Sample EA-561 yields CA-TIMS 206Pb/238U ages
ranging from 38.39 ± 0.07 to 37.49 ± 0.10 Ma. Despite the
relatively small age difference (0.9 m.y.) between the oldest
and youngest zircon grain, the ive ages are not statistically
equivalent (weighted mean age is 38.19 ± 0.32 Ma, MSWD =
57) (Figs. 3A, 5H). This is a typical case showing that LA-ICP-
MS weighted mean dates are often older than the youngest
CA-TIMS age (von Quadt et al., 2011; Chiaradia et al., 2013,
2014). Multiple populations in LA-ICP-MS dates are difi-
cult to interpret, and a consequence is that the LA-ICP-MS
weighted mean ages, in some cases, may not be geologically
meaningful (von Quadt et al., 2011; Chiaradia et al., 2013,
2014; Chelle-Michou et al., 2014). Assuming that chemical
abrasion performed in the high-precision CA-TIMS method
allows complete removal of the effect of postcrystallization Pb
loss (damaged zircon domains) (Mattinson, 2005), the young-
est zircon grains might best approximate the emplacement
age (maximum crystallization age) resolving the 206Pb/238U
LA-ICP-MS data ambiguities. Therefore, the best estimated
age for the dacite porphyry is 37.49 ± 0.10 Ma (youngest
grain, CA-TIMS). This age is identical (within uncertainty)
to 206Pb/238U CA-TIMS age of the equigranular, mesocratic
Clara granodiorite (37.45 ± 0.09 Ma, sample 1116), but older
than the CA-TIMS age for the rhyolite porphyry (Table 3,
Fig. 3A). This proves that at least the early generations of por-
phyry dikes, such as the dacite units (sample EA-561), were
essentially coeval with some of the coarse-grained, equigranu-
lar intrusive rocks seated directly below (sample EA-1116),
which may be part of a larger magmatic reservoir at depth.
Rhyolite porphyry (EA-1284)
This sample comes from a nearly vertical dike intruding
the Clara granodiorite. It is composed of abundant quartz
eyes, plagioclase, and minor biotite phenocrysts in a micro-
felsic groundmass. It is a leucocratic, porphyritic to locally
16 CORREA ET AL.
glomerophyric rock (Fig. 4I, J) dominated by subhedral,
concentrically zoned plagioclase phenocrysts (46.86 vol %).
Microtexture analysis (Table 2) shows large (1,400 µm) phe-
nocryst cores with anorthite compositions ranging from An30
to An21, Sr contents from 1,100 to 1,700 ppm, and Fe con-
tents from 700 to 1,380 ppm. Some grains exhibit multiple
resorption surfaces, with anorthite contents ranging from
An42 to An35, Sr contents from 1,800–2,000 ppm, and Fe
contents from 900–1,500  ppm. These surfaces are followed
by ine-scale oscillatory zoning with compositions similar to
the core. Medium-sized (900–1,000  µm) phenocrysts pres-
ent rounded cores and wavy, ine-scale oscillation zoning with
anorthite content ranging between An35 and An23, Sr between
1,699 and 1,193 ppm, and Fe between 1,381 and 880 ppm.
Toward the rims, anorthite contents range from An12 to An19,
Sr <800 ppm, and Fe <700 ppm. Other phenocrysts include
subhedral K-feldspar (11.04 vol %) with perthitic texture,
anhedral subrounded quartz crystals (8.36 vol %) with embay-
ments, and tabular biotite (1 vol %). Phenocrysts sizes range
between 300 and 800  µm, while groundmass grain size is
restricted to 30 to 120 µm. The groundmass is composed of
quartz (12.96 vol %) and K-feldspar (11.35 vol %), minor pla-
gioclase (4.58 vol %) and scarce, ine-grained biotite altered
to chlorite (2 vol %). Magnetite, titanite, apatite, and zircon
occur as accessory phases. Zircons are less abundant than in
other unit samples, are subhedral to euhedral, and are found
both in the groundmass and as inclusions in plagioclase phe-
nocrysts. The average size is 35 µm, similar to those in the
dacite porphyry, but smaller than the zircons of equigranular
rocks. Biotite, chlorite, and epidote are present as alteration
minerals. The mineralization in this sample is associated with
thin A- and AB-type veinlets with chalcopyrite and K-feldspar
halos.
The LA-ICP-MS weighted mean 206Pb/238U age for this
sample is 38.48 ± 0.62 Ma (MSWD = 1.7) (Fig. 5I). For the
weighted mean 206Pb/238U age calculation (Fig. 3B), two con-
cordant zircons (Fig. 5 I, J) with slightly older ages (42.90 ±
0.62 and 51.50 ± 1.18 Ma) were rejected (black bars, Fig. 3B,
digital supp.). The weighted mean age considering these two
zircons becomes 39.1 ± 1.3 Ma (MSWD = 8.7). The removed
zircons are interpreted as inherited grains. Additionally, a
Middle Triassic (236.9 ± 2.1 Ma) inherited zircon core was
also removed (digital supp.). Five grains were dated by CA-
TIMS, two of which are older inherited zircons (126.95 ±
0.27  Ma, Early Cretaceous, and 202.35 ± 0.30 Ma, Juras-
sic). The remaining three 206Pb/238U CA-TIMS dates are not
statistically equivalent (weighted mean age is 37.1 ± 1.9 Ma,
MSWD = 359), and show a wide age dispersion (~2.5 m.y.)
ranging from 38.58 ± 0.13 to 36.07 ± 0.15 Ma (Figs. 3A, 5J).
Considering the removal of zircon damage zones by the CA-
TIMS method in the same way as for the dacite porphyry,
the youngest zircon (36.07 ± 0.15 Ma, CA-TIMS) is the best
estimated age for the rhyolite porphyry intrusion. Again this
is a demonstration of the fact that weighted mean LA-ICPMS
ages are often systematically older than the youngest CA-
TIMS ages. The 206Pb/238U CA-TIMS best estimated intru-
sion ages of both porphyry samples (37.49 ± 0.10 and 36.07
± 0.15 Ma) are younger than the best estimated 206Pb/238U
CA-TIMS age for the Clara granodiorite (38.71 ± 0.11 Ma), a
result consistent with ield observations.
Subtle inheritance
The presence of (concordant) zircons slightly older than the
crystallization age of the intrusion being dated has been called
subtle inheritance (Campbell et al., 2006). Recycling of these
zircon antecrysts during successive magmatic injections is the
primary cause of the modest age dispersion of concordant
zircon ages, and is compatible with progressive growth of a
large, long-lived, crystal mush body (Miller et al., 2007). The
inclusion of these inherited grains in the calculation of the
LA-ICP-MS weighted mean age may produce high values of
MSWD (>2), suggesting that the spread of ages is not within
analytical uncertainty and thus it could have geologic mean-
ing (Campbell et al., 2006; Chiaradia et al., 2014). Despite
the statistical processing to remove inherited grains and
grains with Pb loss, the ages determined by lower precision
(1–3%) LA-ICP-MS tend to be older than those obtainable by
measuring single zircon grains by higher precision (≤0.1%)
CA-TIMS, implying that inferring an intrusion emplacement
age by in situ methods may be erroneous (von Quadt et al.,
2011; Chiaradia et al., 2014). In this context, the CA-TIMS
age of the youngest single zircon grain is considered the one
that most closely approximates the real timing of intrusion
emplacement (Chiaradia et al., 2013, 2014). Additionally, a
number of recent studies carried out on intermediate and fel-
sic magmatic intrusions show that CA-TIMS age of individual
zircon crystals from the same hand-sized sample are not over-
lapping in age, which cannot be attributed to Pb loss (all zir-
con treated with chemical abrasion). It is rather interpreted
to be the consequence of zircon grains recording a protracted
evolution of magmas within the crust (von Quadt et al., 2011;
Chiaradia et al., 2013, 2014 and references therein).
In this work, an important spread in 206Pb/238U LA-ICP-MS
and CA-TIMS dates (beyond analytical uncertainty) is docu-
mented (Fig. 3A, B). The systematic mixing of old inherited
zircon cores with younger Eocene overgrowths cannot account
for the range of 206Pb/238U dates (see App. 1, 3). The time
elapsed since emplacement until the magmatic body is cooled
to the solidus can neither reproduce the observed zircon age
variability. For example, if the amphibole-rich variety of the
Clara granodiorite (sample EA-1116) is considered, the 1.02 ±
0.12 Ma (2σ) CA-TIMS age spread is incompatible with the geo-
logical evidence. This sample represents a thin band (~0.5 m)
with gradual contacts toward a more typical Clara granodiorite.
Such a range (e.g., 1.02 ± 0.12 Ma in the amphibole-rich variety
of the Clara granodiorite, sample EA-1116) most likely derives
from zircon recycling from older units—i.e., already consoli-
dated parts of the large Clara granodiorite, and perhaps also
partial assimilation of the Equis quartz-monzodiorite (Fig. 7).
We speculate that this “cannibalization” of early-formed
rocks by later magmatic activity (subtle inheritance of zircons)
was probably associated with building of a magmatic reservoir
from which porphyritic dike intrusions associated with miner-
alization were fed (e.g., Chiaradia et al., 2013, 2014). This may
have resulted from the intrusion of a hotter, slightly less evolved
magma, as evidenced by the coeval dacite porphyry event, with
SiO2 contents lower than the typical Clara granodiorite (Fig.
7). Additional evidence of a thermal input during this canni-
balization stage can be found in microtextures preserved in
plagioclase phenocrysts from porphyry dike samples, including
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
the presence of plagioclase core dissolution textures, patchy
textures, and sieve textures as summarized in Table 2. These
microtexture features and compositional variation of plagio-
clase (Table 2) are interpreted as evidence of interaction with a
hot melt (e.g., Pietranik et al., 2006; Renjith, 2014). The large
CA-TIMS age spread in the El Abra dacite (892 ± 122 ka, 2σ)
and rhyolite (2.50 ± 0.20 Ma, 2σ) porphyry dikes, along with
their small sizes (typically narrower than ~80 m) supports the
idea of zircon recycling as a result of fresh magma input provid-
ing additional heat for assimilation of earlier intrusives.
All rejected 206Pb/238U LA-ICP-MS and CA-TIMS con-
cordant data in the range of ~42 to 45 Ma may represent a
mixing of autocrystic and antecrystic (slightly older) zircons
(e.g., Miller et al., 2007) (subtle inheritance) evidencing a
protracted evolution of the El Abra magmas.
Chronological history of the magmatic activity
We have documented protracted magmatic activity at the El
Abra deposit. It extends back until ~45 Ma, as evidenced by
a concordant zircon grain of this age (44.70 ± 0.53 Ma; LA-
ICP-MS, Fig. 7) from the dacite porphyry (sample EA-561),
despite the fact that there is no surface evidence for such
early magmatic activity. After a ~1.8 m.y. magmatic lull, igne-
ous activity resumed at about 43 Ma with the emplacement of
Pajonal diorite (42.9 ± 0.5 to 43.21 ± 0.21 Ma; Table 1). The
zircon record of magmatic activity for the next 2 Ma, from
about 43 and 41 Ma, is scarce for surface outcrops (Fig. 7),
but is documented by slightly older (antecrystic) zircons in
Rojo Grande granite, dacite, and rhyolite porphyries (Fig.
7). The next phase of igneous activity is indicated by ~41 to
36 Ma zircon U-Pb dates (Fig. 7). During this new magmatic
cycle the El Abra granodiorite complex was emplaced, start-
ing at 40.83 ± 0.34 Ma with the earliest phases of Central and
Dark diorites, and ending with the rhyolite dikes at 36.07 ±
0.15 Ma (Fig. 7). With few exceptions, the early magmatic
period, from ~45 until ~37.5 Ma, was dominated by equi-
granular intrusive rocks with relatively low La/Yb and Sr/Y
ratios (Ballard et al., 2001; Fig. 7). In contrast, the late and
more focused igneous activity, from ~37.5 until ~36.1 Ma
(~1.4 m.y.), was exclusively porphyritic at present exposure
levels, with a geochemical signature characterized by high La/
Yb and Sr/Y ratios (Ballard et al., 2001; Fig. 7). At the open
pit mine, an 8.63 ± 0.55-m.y., long-lived magmatic system is
thus recognized.
At the district scale (Fig. 8), porphyry-style magmatism is
almost coeval with the porphyritic magmatism at El Abra. It
was dated by the LA-ICP-MS method in at least three pros-
pects deining a linear array, which includes Brujulina (37.5
± 0.4 and 38.3 ± 0.9 Ma), Cerro Las Papas (also known as
Pinorca) (37.2 ± 0.5 to 34.3 ± 0.4 Ma), and Los Volcanes (36.8
± 0.5 to 35.2 ± 0.5 Ma) (Perelló et al., 2012; Rojas, 2015).
Therefore, at the district scale, igneous activity extended until
34.3 ± 0.5 Ma, making the magmatic history even longer, up
to 10.40 ± 0.73 m.y.
The early magmatic period, between ~45 and 43.2 Ma is
about 2 m.y. older than previously recognized at the El Abra
deposit, and may represent the earliest phase of the Pajonal
complex. Another alternative is a group of rocks without a for-
mational name (quartz monzodiorite to diorite), which crop
out ~5 km northeast of El Abra and are reported to have an
17
age of 45.5 Ma (whole-rock K/Ar; Tomlinson et al., 2001).
Similarly, ages of about 45 Ma have also been reported for
the Los Picos quartz monzodiorite, located at present ~35 km
southward from El Abra and westward from the West fault,
in the Chuquicamata district (Fig. 8). The Los Picos unit is
broadly correlative with the Pajonal diorites and was spatially
associated with the El Abra deposit in pre-Oligocene times
(e.g., Dilles et al., 1997; see below).
Finally, magmatism at the El Abra deposit has evolved
within (and assimilated) Mesozoic crustal rocks at depth,
as indicated by xenocrystic zircons with ages ranging from
102.07 ± 0.15 to 236.9 ± 2.1 Ma (Table 3).
Results and Discussion: Mineralization
Age of ore formation: Re-Os molybdenite geochronology
The concentrations of Re and Os in molybdenite and calcu-
lated ages are summarized in Table 4. The total Re and 187Os
concentrations range from 242.8 to 508.2 ppm and from 92 to
193.4 ppb, respectively. The highest concentrations of Re and
Os are reported in sample EA-916, located in the northeast
part of the mine (Fig. 1), corresponding to a D-type veinlet
within the dacite variety of the El Abra porphyry. The age is
36.34 ± 0.18 Ma (2σ uncertainty). Samples EA-903 (36.18 ±
0.18 Ma) and EA-710 (36.22 ± 0.18 Ma) (belonging to B-type
veinlets), hosted by the dacite and rhyolite porphyry respec-
tively, have lower Re abundances and are within the 2σ uncer-
tainty of the age of sample EA-916.
Three Re-Os dates were obtained from molybdenite sam-
ples that spread over a proile of ~1.3 km (Fig. 1), and from
different veinlet (B and D) and rock types (rhyolite and dacite
porphyries). Nevertheless, the Re-Os data deine a tight clus-
ter, between 36.34 ± 0.18 and 36.18 ± 0.18 Ma. These high-
precision Re-Os dates are in agreement with previous works,
which (based on K-Ar data of potassic alteration) suggest a
cooling window for the hydrothermal luids from about 36.8
to 35.0 Ma (Ambrus, 1977; recalculated by Maksaev, 1990;
Dilles et al., 1997; Tomlinson et al., 2001). According to these
new data, the ore formation age (mean: 36.25 ± 0.25 Ma, 2σ)
is identical within uncertainty, to the age of the latest mag-
matic activity occurred in the deposit site represented by the
rhyolite porphyry (36.07 ± 0.15 Ma).
The geologic evidence at the El Abra deposit suggests
multiple events of porphyry intrusion lasting for ~1.4 m.y.
and multiple pulses of veining (Fig. 2B), but the statistically
equivalent Re-Os dates constrain the mineralizing pulses to
a focused episodes over a 0.16 ± 0.25 m.y. period, consider-
ing the upper and lower bounds of uncertainty of the Re-Os
ages. Even if these Re-Os results could be the outcome of
undersampling, it does not downplay the signiicance of the
widespread ore-forming event at 36.25 ± 0.25 Ma (2σ).
At the Los Volcanes porphyry Cu-Mo prospect, in the El
Abra district (Fig. 9), Re-Os geochronology indicates that
mineralization was coeval with that of the El Abra deposit,
despite being located ~10 km away from it. In this pros-
pect, the Re-Os dates deine a series of mineralization pulses
spanning 1.02 ± 0.21 m.y. (2σ), from 36.47 ± 0.15 to 35.45 ±
0.14 Ma (Perelló et al., 2012; Rojas, 2015).
Together, ield observations and the U-Pb and Re-Os geo-
chronology data reveal a cluster of magmatic-hydrothermal
18 CORREA ET AL.

E 535.000
E 525.000
E 505.000

E 515.000
N 7.580.000

El Abra Cluster

Brujulina El Abra Los Volcanes

(Conchi)

Cerro Las Papas

(Pinorca) N 7.570.000

Eocene location of
Toki Cluster
N 7.560.000 EOCENE COMPLEXES
Pajonal-El Abra Complex

Pajonal diorite
(42.9±0.5 - 43.21±0.21 Ma)

El Abra granodiorites:
-Central diorite (40.62±0.31 Ma)
-Equis qtz-monzonite
N 7.550.000 (41.74±0.19 - 38.5±0.4 Ma)

Llareta granodiorite
(40.10± 0.48 Ma)

Clara granodiorite
Radomiro Tomic (38.74±0.10 - 37.55±0.20 Ma)

Los Picos-Fortuna Complex

N 7.540.000 Los Picos monzodiorite complex
(41.5-43.7 Ma)

Fortuna granodioritic complex

Chuquicamata -Tetera aplitic porphyries
-San Lorenzo granodiorite and
Tonalite porphyries (38.2 ± 0.8 Ma)
-Fiesta granodiorite (36.9 - 39.9 Ma)
-Antena granodiorite (39.3±0.4 Ma)
Mina Sur N 7.530.000
Chuquicamata Porphyry Complex
East porphyry (35.2±0.4 Ma)
West porphyry (34±0.3 Ma)
Ministro Hales Banco porphyry (34.1±0.3 Ma)
Toki Cluster
Other units
Q T Opache prospect
O
G
M N 7.520.000
G Genoveva prospect
Q Quetena prospect
O
Calama M Miranda prospect
T Toki prospect

Fig. 8. Location map for the principal deposits and prospects of the El Abra and Chuquicamata districts, showing the Toki
cluster position at 37 to 39 Ma following the 35-km left lateral offset of the West fault (after Dilles et al., 1997). Projection:
UTM19; Datum: Provisional South America 1956 (PSAD56). Modiied from Barra et al. (2013) and Barrett (2004).
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
Fig. 9. Location and ages of prospects in the El Abra cluster. Ages in black = U-Pb ages of porphyry units, Ages in blue =
Re-Os in Mo ages. Modiied from Tomlinson et al. (2001).
centers in the El Abra district (Fig. 9), pointing to the pres-
ence of a large parental reservoir to which these shallow-level
porphyry systems are linked, which makes the El Abra district
attractive for further ore exploration at the district scale.
Regional correlation along the Domeyko fault system
The hypothesis of post-Oligocene sinistral displacement
along the West fault, a structure related to the Domeyko
fault system, has been widely studied and conirmed by the
petrographic and geochronological correlation between the
Los Picos-Fortuna and the Pajonal-El Abra intrusive com-
plexes (Reutter et al., 1996; Dilles et al., 1997; Tomlinson et
al., 2001; Ballard, 2001; Ballard et al., 2002; Campbell et al.,
2006; among others).
The Re-Os geochronological data obtained in this study and
those reported from Los Volcanes (Perelló et al., 2012) are
similar to the Re-Os mineralization ages reported by Barra
et al. (2013) for Miranda (36.5 ± 0.2 to 36.0 ± 0.2 Ma) and
19
Opache (37.6 ± 0.2 to 36.4 ± 0.2 Ma), the youngest porphyry
copper deposits from the Toki cluster (Rivera and Pardo,
2004), which is located about 50 km southward of the El Abra
deposit, in the Chuquicamata area (Fig. 8). In this cluster, the
nominally older deposits of Quetena, Toki, and Genoveva have
Re-Os ages at around 38 Ma (Barra et al., 2013). Together,
the ore formation ages of the ive porphyry Cu-Mo deposits
display a ~2.4 m.y. spread, from 38.4 ± 0.2 to 36.0 ± 0.2 Ma
(Barra et al., 2013). Inferred resources for the Toki cluster are
estimated at about 20 Mt of Cu (Camus, 2003; Rivera et al.,
2006).
When tectonic reconstruction of the ~35-km offset along
the Domeyko fault system is considered, then the location
of the Toki cluster at 37 to 39 Ma is ~15 km south from the
present position of the El Abra region (Fig. 8). This scenario
suggests that during the late Eocene both clusters might have
formed a single, large (~30 km across) porphyry copper cen-
ter associated with the youngest porphyritic facies from the
20 CORREA ET AL.
Pajonal-El Abra-Los Picos-Fortuna batholith. Subsequently,
the “El Abra-Toki” porphyry copper cluster (and batholith)
containing resources estimated at ~28.5 Mt of Cu (Camus,
2003; Rivera et al., 2006) was tectonically bisected and the
individual parts were displaced to their present positions (Fig.
8) by post-Eocene left-lateral movements along the same
strike-slip structural system that controlled their formation, as
depicted by earlier tectonic models (e.g., Dilles et al., 1997,
and references therein).
Acknowledgments
This work is part of a PhD study of the irst author, at the Uni-
versidad de Concepción. We thank the Sociedad Contractual
Minera El Abra (SCMEA) and Freeport-McMoRan Inc., for
funding this research, for the logistical support during ield
work of the irst author, and inally for the permission to pub-
lish the results. We also thank Ricardo Ardila for his coop-
eration during ield work activities and Richard Friedman for
the LA-ICP-MS and CA-TIMS zircon analyses. Stephanie
Mrozek and Massimo Chiaradia are thanked for thorough
reviews and constructive comments that helped to improve
this paper.
REFERENCES
Ambrus, J., 1977, Geology of the El Abra porphyry copper deposit, Chile:
Economic Geology, v. 72, p. 1062–1085.
——1979, Emplazamiento y mineralización de los póridos cupríferos de
Chile: Unpublished Ph.D. thesis, España, Universidad de Salamanca,
314 p.
Ardila, R., 2009, Eventos de mineralización de sulfuros en la mina El Abra:
Internal report from Sociedad Contractual Minera El Abra (SCMEA), 54 p.
Ballard, J.R., 2001, A comparative study between the geochemistry of ore-
bearing and barren calc-alkaline intrusions: Unpublished PhD thesis, Can-
berra, Australia, The Australian National University, 254 p.
Ballard, J.R., Palin, J.M., Williams, I.S., Campbell, I.H., and Faunes, A.,
2001, Two ages of porphyry intrusion resolved for the super-giant Chuqui-
camata copper deposit of northern Chile by ELA-ICP-MS and SHRIMP:
Geology, v. 29, p. 383–386.
Ballard, J., Palin, M., and Campbell, I., 2002, Relative oxidation states of
magmas inferred from Ce(IV)/Ce(III) in zircon: application to porphyry
copper deposits of northern Chile: Contributions to Mineralogy and Petrol-
ogy, v. 144, p. 347–364.
Barra, F., Alcota, H., Rivera, S., Valencia, V., Munizaga, F., and Maksaev,
V., 2013, Timing and formation of porphyry Cu-Mo mineralization in the
Chuquicamata district, northern Chile: New constraints from the Toki clus-
ter: Mineralium Deposita, v. 48, p. 629–651.
Barrett, L.F., 2004, Geology and status of exploration projects, El Abra dis-
trict, II Region, Chile, El Abra: Sociedad Contractual Minera El Abra
(SCMEA), Unpublished report, p. 6–37.
Campbell, I.H., Ballard, J.R., Palin, J.M., Allen, C., and Faunes, A., 2006,
U-Pb zircon geochronology of granitic rocks from the Chuquicamata-El
Abra porphyry copper belt of northern Chile: Excimer laser ablation ICP-
MS analysis: Economic Geology, v. 101, p. 1327–1344.
Camus, F., 2003, Geología de los sistemas porfíricos en los Andes de Chile:
Corporación Nacional del Cobre de Chile-Servicio de Geología y Minería
de Chile, 267 p.
Chelle-Michou, C., Chiaradia, M., Ovtcharova, M., Ulianov, A., and Wotzlaw,
J.F., 2014, Zircon petrochronology reveals the temporal link between por-
phyry systems and the magmatic evolution of their hidden plutonic roots
(the Eocene Coroccohuayco deposit, Peru): Lithos, v. 198–199, p. 129–140.
Chiaradia, M., Schaltegger, U., Spikings, R., Wotzlaw, J.F., and Ovtcharova,
M., 2013, How accurately can we date the duration of magmatic-hydrother-
mal events in porphyry systems?: Economic Geology, v. 108, p. 565–584.
Chiaradia, M., Schaltegger, U., and Spikings, R.A., 2014, Time scales of min-
eral systems—advances in understanding over the past decade: Society of
Economic Geologists, Special Publication 18, p. 37–58.
Coira, B., Davidson, J., Mpodozis, C., and Ramos, V., 1982, Tectonic and
magmatic evolution of the Andes of northern Argentina and Chile, in
Linares E., ed., Magmatic evolution of the Andes, Earth Science Reviews,
Special Issue 18, p. 303–332.
Dilles, J.H., Tomlinson, A., Marti, M.W., and Blanco, N., 1997, El Abra
and Fortuna Complexes: A porphyry copper batholith sinistrally dis-
placed by the Falla Oeste: Actas VIII, Congreso Geológico Chileno, v. 3,
p. 1883–1887.
Gerstenberger, H., and Haase, G., 1997, A highly effective emitter substance
for mass spectrometric Pb isotopic ratio determinations: Chemical Geology,
v. 136, p. 309–312.
Grifin, W.L., Powell, W.J., Pearson, N.J., and O’Reilly, S.Y., 2008, Glitter:
Data reduction software for laser ablation ICP-MS, in Sylvester, P.J., ed.,
Laser ablation ICP-MS in the earth sciences: Current practices and out-
standing issues: Vancouver, B.C., Mineralogical Association of Canada
Short Course Series, Short Course 40, p. 308–311.
Haschke, M.R., 2002, Evolutionary geochemical patterns of Late Cretaceous
to Eocene arc magmatic rocks in north Chile: Implications for Archean
crustal growth: European Geosciences Union, EGU Stephan Mueller Spe-
cial Publication Series, v. 2, p. 207–218.
Jaffey, A.H., Flynn, K.F., Glendenin, L.E., Bentley, W.C., and Essling, A.M.,
1971, Precision measurement of half-lives and speciic activities of 235U and
238U: Physical Review Letters, v. 4, p. 1889–1906.
Jarosewich, E., Nelen, J.A., and Norbery, J.A., 1980, Reference samples for
electron microprobe analysis: Geostandards Newsletter, no. 4, p. 43–47.
Ludwig, K.R., 2003, Isoplot/Ex. Version 3: A geochronological toolkit for
Microsoft Excel, Berkeley Geochronology Center, California.
Maksaev, V., 1990, Metallogeny, geological evolution and thermocronology of
the Chilean Andes between latitudes 21° and 26° south, and the origin of
major porphyry copper deposits: Unpublished PhD thesis, Halifax, Canada,
Dalhousie University, 544 p.
Mattinson, J.M., 2005, Zircon U-Pb chemical abrasion (“CA-TIMS”) method:
Combined annealing and multi-step partial dissolution analysis for improved
precision and accuracy of zircon ages: Chemical Geology, v. 220, p. 47–66.
Miller, J.S., Matzel, J.E.P., Miller, C.F., Burgess, S.D., and Miller, R.B.,
2007, Zircon growth and recycling during the assembly of large, compos-
ite arc plutons: Journal of Volcanology and Geothermal Research, v. 167,
p. 282−299.
Mpodozis, C., Marinovic, N., Smoje, I., and Coutiño, L., 1993, Estudio
geológico estructural de la cordillera de Domeyko entre Sierra Limón
Verde y Sierra Mariposas, región de Antofagasta: Servicio Nacional de
Geología y Minería-Corporación Nacional del Cobre (Chile), Informe reg-
istrado IR-93-04, 282 p.
Mundil, R., Ludwig, K.R., Metcalfe, I., and Renne, P.R., 2004, Age and tim-
ing of the Permian mass extinctions: U/Pb dating of closed-system zircons:
Science, v. 305, p. 1760–1763.
Perelló, J., Saldías, G., Gonzales, M., Hernríquez, R., and Giglio, S., 2012,
Hypogene alteration and mineralization at the Conchi porphyry Cu-Mo
deposit, northern Chile: assemblages, zoning, and age: XVI Peruvian Geo-
logic Congress, Lima, 2 p.
Pietranik, A., Koepke, J., and Puziewicz, J., 2006, Crystallization and resorp-
tion in plutonic plagioclase: Implications on the evolution of granodiorite
magma (Gȩsiniec granodiorite, Strzelin Crystalline Massif, SW Poland):
Lithos, v. 86, p. 260–280.
Renjith, M.L., 2014, Micro-textures in plagioclase from 1994–1995 eruption,
Barren Island Volcano: Evidence of dynamic magma plumbing system in
the Andaman subduction zone: Geoscience Frontiers, v. 5, p. 113–126.
Reutter, K.J., Scheuber, E., and Chong, G., 1996, The Precordilleran fault
system of Chuquicamata, northern Chile: evidence for reversals along arc-
parallel strike-slip faults: Tectonophysics, v. 259, p. 213–228.
Rivera, S.L., and Pardo, R., 2004, Discovery and geology of the Toki por-
phyry copper deposit, Chuquicamata district, northern Chile: Society of
Economic Geologists, Special Publication 11, p. 15–54.
Rivera, S.l., Pardo, R., Alcota, H., Fontecilla, C., Kovacic, P., Pizarro, J., and
Rojo, J., 2006, Reseña de la exploración y geología del cluster de póridos de
cobre Toki, distrito Chuquicamata: Actas XI Congreso Geológico Chileno,
Antofagasta (electronic version), p. 343–346.
Rojas, G., 2015, Petrografía, geoquímica y geocronología del distrito Los Vol-
canes (Conchi), Región de Antofagasta, Chile: Unpublished thesis, Chile,
Universidad de Concepción, 139 p.
Santana, C.A., 2010, Informe estudio geológico-estructural y su aplicación
a la exploración de cuerpos mineralizados en el distrito El Abra: Internal
report for El Abra Company (SCMEA), 25 p.
Schmitz, M.D., and Schoene, B., 2007, Derivation of isotope ratios, errors, and
error correlations for U-Pb geochronology using 205Pb-235U-(233U)-spiked
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE
isotope dilution thermal ionization mass spectrometric data: Geochemistry,
Geophysics, Geosystems, v. 8, p. 1–20.
Scoates, J.S., and Friedman, R.M., 2008, Precise age of the platiniferous
Merensky reef, Bushveld Complex, South Africa, by the U-Pb ID-TIMS
chemical abrasion ID-TIMS technique: Economic Geology, v. 103,
p. 465–471.
Selby, D., and Creaser, R.A., 2004, Macroscale NTIMS and microscale
LA-MC-ICP-MS Re-Os isotopic analysis of molybdenite: Testing spatial
restrictions for reliable Re-Os age determinations, and implications for the
decoupling of Re and Os within molybdenite: Geochimica et Cosmochi-
mica Acta, v. 68, p. 3897–3908.
Selby, D., Creaser, R.A., Stein, H. J., Markey, R. J., and Hannah, J. L., 2007.
Assessment of the 187Re decay constant accuracy and precision: Cross
calibration of the 187Re-187Os molybdenite and U-Pb zircon chronometers:
Geochimica et Cosmochimica Acta, v. 71, 1999–2013.
Sláma, J., Košler, J., Condon, D.J., Crowley, J.L., Gerdes, A., Hanchar, J.M.,
Horstwood, M.S.A., Morris, G.A., Nasdala, L., Norberg, N., Schaltegger,
U., Xchoene, B., Tubrett, M.N., and Whitehouse, M.J., 2007, Plešovice
zircon—a new natural reference material for U–Pb and Hf isotopic micro-
analysis: Chemical Geology, v. 249, p. 1–35.
Smoliar, M.I., Walker, R.J., and Morgan, J.W., 1996, Re-Os isotope con-
straints on the age of Group IIA, IIIA, IVA, and IVB iron meteorites: Sci-
ence, v. 271, p. 1099–1102.
21
Tafti, R., Mortensen, J.K., Lang, J.R., Rebagliati, M., and Oliver, J.L., 2009,
Jurassic U-Pb and Re-Os ages for newly discovered Xietongmen Cu-Au
porphyry district, Tibet: Implications for metallogenic epochs in the south-
ern Gangdese belt: Economic Geology, v. 104, p. 127–136.
Thirlwall, M.F., 2000, Inter-laboratory and other errors in Pb isotope analyses
investigated using a 207Pb-204Pb double spike: Chemical Geology, v. 163,
p. 299–322.
Tomlinson, A.J., Blanco, N., Maksaev, V., Dilles, J.H., Grunder, A.L., and
Ladino, M., 2001, Geología de la precordillera Andina de Quebrada
Blanca- Chuquicamata, Regiones I y II (20°30'-22°30' S): Servicio Nacional
de Geología y Minería (Chile), Informe Registrado IR-01-20, 444 p.
Valente, D.L., Campbell, I.H., and Allen, C.M., 2006, The geology, geochem-
istry and geochronology of the El Abra mine, Chile and the adjacent El
Abra-Pajonal suite [abs.]: Geochimica et Cosmochimica Acta, v. 70, p. A664.
van Achterbergh, E., Ryan, C.G., Jackson, S.E., and Grifin, W.L., 2001,
Data reduction software for LA-ICP-MS: Appendix; in Sylvester, P.J., ed.,
Laser ablation–ICP-mass spectrometry in the earth sciences: Principles
and applications: Mineralogical Association of Canada Short Course Series,
Ottawa, Ontario, Canada, v. 29, p. 239–243.
von Quadt, A., Emi, M., Martinek, K., Moll, M., Peycheva, and Heinrich,
C.A., 2011, Zircons crystallization and the lifetimes of ore-forming mag-
matic-hydrothermal systems: Geology, v. 39, p. 731–734.
22 CORREA ET AL.

APPENDIX 1
CL Images of Dated Zircons from Different Rock Types

1 2 3 6
8 9

4 5
15 16 17
11 12

13 18 19
14

20 21 22

23
24

App. 1, Fig. 1. Examples of CL images from dated zircons: 1-5. Rojo Grande granite (EA-657); 6-9. Dacite porphyry
(EA-561); 11-14. Mesocratic Clara granodiorite (EA-1116); 15-19. Rhyolite porphyry (EA-1284); 20-24. Clara granodiorite
(EA-1141).
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE 23

APPENDIX 2
Apatite Inclusions in Zircon

A B

C D

A B
D
App. 2, Fig. 1. Images of apatite inclusions in zircons. A. BSE. B. SEI. C. BSE. D. Ca Kα X-ray map.

App. 2, Table 1. Apatite Composition (EDS analysis spot 74

App. 2, Fig. 1C, above)

Spectrum: 74
unn. C norm. C Atom. C Error (1σ)
El AN Series [wt %] [wt %] [at.%] [w %]

O 8 K-series 17.76 23.78 41.62 2.57

F 9 K-series 1.29 1.72 2.54 0.33
P 15 K-series 13.34 17.86 16.14 0.54
Cl 17 K-series 0.99 1.32 1.05 0.06
Ca 20 K-series 41.33 55.32 38.65 1.26

Total: 74.71 100.00 100.00

24 CORREA ET AL.

A B

App. 2, Fig. 2. Images of apatite inclusions and trace element-rich zones in zircons. A. BSE. B. SEI. C. BSE.

App. 2, Table 2. Apatite Composition (EDS analysis spots

App. 2, Fig. 2B, above)

Norm. mass percent (%)

Spectrum O F P Cl Ca

50 23.88 1.05 22.21 2.06 50.79

51 24.28 2.13 19.10 0.69 53.80
52 24.08 0.85 19.09 1.92 54.06

Mean value: 24.08 1.34 20.14 1.56 52.88

Sigma: 0.20 0.69 1.80 0.75 1.82
Sigma mean: 0.11 0.40 1.04 0.43 1.05
 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE 25

cps/eV
2.2 Hf Th O Hf Si Zr
2.0
U
1.8

1.6

1.4

1.2
U
1.0 Th
0.8

0.6

0.4
Hf
0.2
Th U
0.0
2 4 6 8 10 12 14
keV

App. 2, Fig. 3. EDS spectra for the trace element-enriched zone in zircon (spot 80 in image App. 2, Fig. 2C.)

App. 2, Table 3. Compositional Analysis Th, U, Hf

(enriched zone spot 80 in image App. 2, Fig. 2C)

unn. C norm. C Atom. C Error (1σ)

El AN Series [wt %] [wt %] [at.%] [w %]

O 8 K-series 16.22 15.99 41.95 2.00

Si 14 K-series 20.24 19.96 29.83 0.87
Zr 40 L-series 60.08 59.23 27.26 2.26
Hf 72 L-series 1.68 1.65 0.39 0.10
Th 90 M-series 1.60 1.58 0.29 0.08
U 92 M-series 1.61 1.59 0.28 0.08

Total: 101.43 100.00 100.00

26 CORREA ET AL.

A B

App. 2, Fig. 4. Images of apatite inclusions in zircons. A. SEI. B. SEI.

App. 2, Table 4. Apatite Composition (EDS analysis spot 63 in

App. 2, Fig. 4B, above)

Spectrum: 63

unn. C norm. C Atom. C Error (1σ)

El AN Series [wt %] [wt %] [at.%] [w %]

O 8 K-series 19.14 24.21 42.31 2.76

F 9 K-series 1.20 1.52 2.23 0.32
P 15 K-series 13.57 17.17 15.50 0.54
Cl 17 K-series 1.04 1.32 1.04 0.07
Ca 20 K-series 44.09 55.79 38.92 1.35

Totals: 79.04 100.00 100.00

 MAGMATISM AND ORE FORMATION IN THE EL ABRA PORPHYRY Cu DEPOSIT, N. CHILE 27

APPENDIX 3
Th/U Ratio Versus 206Pb/238U CA-TIMS Age Diagrams

To check if the spread in 206Pb/238U CA-TIMS dates observed
in most samples (EA-561, EA-1141, EA-1116, and EA-1284)
may result from mixing different proportions of old inherited
cores and Eocene overgrowths, an evaluation using Th and U
contents in zircons and age information is done. If the presence
of small xenocrystic cores were responsible for the spread in
206Pb/238U CA-TIMS dates, data in a Th/U ratio vs. 206Pb/238U
CA-TIMS age diagram should be plotted along a mixing line
between the old core and the younger overgrowth. Using
the middle Cretaceous age of 102.07 ± 0.15 Ma (206Pb/238U
CA-TIMS) from the Rojo Grande (sample EA-567) as the
old core, the result shows that the 206Pb/238U CA-TIMS dates
from the dacite porphyry (EA-561) zircons do not plot along
the mixing line (App. 3, Fig. 1). If the older early Jurassic
age of 177.6 ± 3.7 Ma (206Pb/238U LA-ICP-MS) is considered
instead, the outcome does not change (App. 3, Fig. 2). Simi-
lar conclusions can be drawn for samples EA-1141, EA-1116,
and EA-1284 (black dots in App. 3, Figs. 1, 2). Therefore, the
presence of small inherited zircon cores cannot adequately
explain the disparity of 206Pb/238U CA-TIMS dates observed
to a sample scale (Fig. 4A).
REFERENCE
Belousova, E.A., Grifin W.L., O’Reilly S.Y., and Fisher, N.I., 2002, Igneous
zircon: Trace element composition as an indicator of source rock type: Con-
tributions to Mineralogy and Petrology, v. 143, p. 602–622.

App. 3, Fig. 1. Th/U vs. 206Pb/238U CA-TIMS diagram showing the theoretical mixing line between the youngest zircon from
the dacite porphyry and the inherited Cretaceous zircon from the Rojo Grande granite. Black dots represent data from other
samples (Clara granodiorite, melanocratic band in Clara granodiorite, and rhyolite porphyry).
28 CORREA ET AL.

0,9
Youngest Dacite porphyry
0,8 zircon (sample EA-561)

0,7 Crustal zircon: Median ~300 ppm U

Green areas >90% crustal Zrn
0,6
50

0,5 100
Th/U

200
0,4
400
0,3 800
1600
3200 Old
0,2
6400
zircon
0,1 core

0,0
30 50 70 90 110 130 150 170 190
206Pb/ 238U age (Ma)
App. 3, Fig. 2. Th/U vs. 206Pb/238U CA-TIMS diagram showing different mixing lines, for different U concentrations, between
the youngest zircon from the dacite porphyry and the inherited Jurassic zircon from the same dacite porphyry and several
mixing lines. Crustal zircon median value estimated after Belousova et al. (2002). Note that the only candidates for old-core
end members (those that have a low enough Th/U ratio to account for the mixing) are the middle Cretaceous (102.07 ±
0.15 Ma; 206Pb/238U CA-TIMS age) zircons from the Rojo Grande granite (sample EA-567) and the early Jurassic (177.6 ±
3.7 Ma; 206Pb/238U LA-ICP-MS age) zircons from the dacite porphyry (sample EA-561). To assess the mixing hypothesis using
the Jurassic zircon data, several mixing lines for different U concentrations were drawn since the U content of this zircon was
not available. Here, the dacite porphyry ages plot along a mixing line, but only at very high U contents (≥5,000 ppm). These
U contents, however, are higher than expected for crustal zircons from granodiorites and tonalites, which are lower than
~3,000 ppm and typically less than ~1,000 ppm (Belousova et al., 2002), making that mixing highly improbable.