JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 6
PAGES 1091^1122
2012
doi:10.1093/petrology/egs010
ZirconTrace Element and O^Hf Isotope Analyses
of Mineralized Intrusions from El Teniente
Ore Deposit, Chilean Andes: Constraints on
the Source and Magmatic Evolution of
Porphyry Cu^Mo Related Magmas
M. MUN‹OZ1*, R. CHARRIER1, C. M. FANNING2, V. MAKSAEV1 AND
K. DECKART1
UNIVERSIDAD DE CHILE, DEPARTAMENTO DE GEOLOGI¤A, PLAZA ERCILLA 803, CASILLA 13518, CORREO 21,
1
SANTIAGO, CHILE
2
RESEARCH SCHOOL OF EARTH SCIENCES, AUSTRALIAN NATIONAL UNIVERSITY, CANBERRA, ACT 0200, AUSTRALIA
RECEIVED DECEMBER 17, 2010; ACCEPTED JANUARY 30, 2012
ADVANCE ACCESS PUBLICATION MARCH 5, 2012
Intrusive rocks related to porphyry copper mineralization are part
of the wide diversity of subduction-related, mantle-derived, igneous
rocks generated in convergent margin settings. What differentiates
them from barren igneous rocks results ultimately from the
multi-component and multi-stage processes that condition magma
composition in these settings. Unfortunately, the petrogenetic history
is largely obscured by the pervasive alteration that affects rocks in
these deposits. We address this issue through the study of zircon
grains from El Teniente, one of the largest known porphyry Cu^Mo
deposits in the world. El Teniente belongs to the Miocene^Pliocene
Cu^Mo belt of the Central Chilean Andes, which formed in a
short timespan during the Cenozoic constructive period of the
orogen. Previously U^Pb dated zircon grains were selected for
re-examination of their morphological characteristics and in situ
analysis of chemical (rare earth element, Hf, Y and Ti contents)
and isotopic (Hf, O) composition. They are from six intermediate
to felsic syn- to late-mineralization, intrusive units covering a timespan of 1·6 Myr. The El Teniente zircons have compositional
and morphological characteristics indicating crystallization from a
series of cogenetic melts. However, a minor hydrothermal imprint
is documented in the presence of crystals with mottled surfaces that
correspond to thin high U^Th overgrowth rims (low-luminescent
features in cathodoluminescence images). In terms of any other chemical and isotopic characteristic, these are indistinguishable from the
main mineral populations. Zircons define morphological and chemical trends reflecting an evolution towards more differentiated
magma compositions, lower crystallization temperatures and
increased cooling rates with decreasing age of intrusion. Hf and O
isotopic compositions are remarkably uniform at grain, sample and
deposit scale. This, together with the general absence of older inherited zircon components, the lack of correlations between isotopic
signature and whole-rock composition and high initial eHf values
(total average 7·4 1·2; 2s), rules out involvement of any significant crustal contamination in the genesis of the El Teniente
magmas. The Hf isotopic composition indicates a relatively juvenile
source, but with some crustal residence time. The d18OZrc weighted
mean of 4·76 0·12ø (2s; 61 analyses) is at the lower limit of
the normal mantle zircon range of 5·3 0·6ø (2s), and might reflect crystallization from low-18O magmas. Hf isotopic compositions
have a restricted range in initial eHf values between þ6 and þ10,
identical to preceding Cenozoic barren magmatic activity in Central
Chile. These igneous rocks are the product of nearly 25 Myr of
subduction-related magmatic activity, developed under contrasting
tectonic regimes and margin configurations.This suggests a primary
control of the isotopic signature by a stable long-lived MASH-type
(melting, assimilation, storage and homogenization) reservoir
in the deep lithosphere. In the context of the Cenozoic evolution of
Central Chile we argue that dehydration melting in the enriched
* Corresponding author. Telephone: 56 2 9784533. Fax: 56 2 6963050.
E-mail: marmunoz@ing.uchile.cl
The Author 2012. Published by Oxford University Press. All
rights reserved. For Permissions, please e-mail: journals.permissions@
oup.com
JOURNAL OF PETROLOGY
VOLUME 53
MASH reservoir occurred as a consequence of increasing crustal
thickness, and was prompted by a high-temperature thermal regime
resulting from long-lasting preceding magmatism. This process can
also fractionate O to generate low-18O magmas. At the time of El
Teniente formation, dehydration melting occurred coevally with arc
migration, which probably influenced the fertility of the magmas by
increasing the melt component derived from this process relative to
the component derived from primary basalt differentiation. At a regional scale, such reactions are expected to occur as a consequence
of progressive crustal thickening during the constructive period of
the Andes, and can explain the simultaneous generation of porphyry
deposits in the Miocene^Pliocene Cu^Mo belt of Central Chile.
El Teniente; porphyry Cu^Mo deposit; zircon; O^Hf
isotopes; dehydration melting
KEY WORDS:
I N T RO D U C T I O N
The relation between porphyry Cu^Mo deposit formation
and active continental margin magmatism has been well
documented (e.g. Lindgren, 1933; Sillitoe, 1972; Burnham,
1979; Cline & Bodnar, 1991; Hedenquist & Lowenstern,
1994). The magmas are the main source of H2O, S, and
Cu, among other elements and compounds, of the hydrothermal systems whose evolution results in the formation
of these ore deposits. However, whereas magmatic and
hydrothermal processes occur widely in subduction-related
arc settings, the formation of large economic porphyry
copper deposits is restricted. These deposits constitute localized chemical and mineralogical anomalies, formed
during a relatively short timespan and at specific moments
during the lifetime of the host magmatic arc (Maksaev &
Zentilli, 1988; McKee & Noble, 1989; Cornejo et al., 1997;
Richards et al., 2001). The many studies performed on
these deposits have well characterized the numerous tectonic, structural, magmatic and chemical conditions optimal for their formation. However, the genesis of magmas
related to porphyry copper mineralization remains a
highly debated issue. Different models have been proposed,
ranging from those invoking primary enrichment owing
to key processes during magma genesis (Kay &
Mpodozis, 2001; Oyarzu¤n et al., 2001; Mungall, 2002; Core
et al., 2006; Shafiei et al., 2009), to those considering them
as the products of the convergence of normal processes
operating in arcs (Richards, 2003, 2005; Stern & Skewes,
2005; Chiaradia et al., 2009; Stern et al., 2010). These
models are not necessarily exclusive, but further research
is needed to understand more accurately the processes
and/or components involved in the genesis of porphyry
copper related magmas at their source, as well as during
their subsequent evolution in their passage through the
upper lithosphere.
The main problem in studying intrusive rocks in such
deposits is the widespread pervasive hydrothermal
NUMBER 6
JUNE 2012
alteration that has modified their primary textural and
chemical characteristics. This has greatly restricted reliable
characterization of geochemical and isotopic primary
signatures by conventional methods. However, the study
of single minerals, made possible by the development of
microanalytical techniques over the past few decades,
allows an insight into the primary characteristics of heavily altered rocks by selectively avoiding the effects of
whole-rock alteration. Zircon is a common accessory mineral particularly well suited for this kind of study. The
physical and chemical stability of zircon, its resistance to
high-temperature diffusive re-equilibration (Watson &
Cherniak, 1997; Cherniak & Watson, 2003), and its tendency to incorporate numerous trace and radiogenic elements make it an ideal mineral to see through the
subsequent alteration commonly seen on the whole-rock
scale. Additionally, zircon is abundant in intermediate to
felsic igneous rocks such as those related to porphyry
copper deposits, making it a valuable tool to track and
characterize the petrogenetic evolution of the magmas
from which it crystallized.
The Chilean continental margin is ideal to examine different aspects of porphyry copper systems as it hosts numerous deposits of this type. Its geological evolution has
been largely linked to abundant intrusive and volcanic activity as a consequence of plate convergence. The current
volcanic arc is located along the axis of the main range
and is represented by the Chilean Northern and Southern
volcanic zones (Fig. 1). Trenchwards, igneous rocks cropping out in different north^south-trending belts are the
remnants of several arcs developed along the continental
margin since Paleozoic times (Mpodozis & Ramos, 1989;
Charrier et al., 2007), some of which host numerous porphyry copper deposits. Their formation is restricted to a
short time interval, during the last stages of the related
arc lifespan, characterized by the waning of widespread intrusive and volcanic activity in a regime of crustal shortening, thickening, and uplifting (for reviews see Camus,
2003; Maksaev et al., 2007; Charrier et al., 2009).
The El Teniente porphyry Cu^Mo deposit is the youngest known deposit of this type within the Chilean continental margin and one of the largest in the world, with 93·7
Mt (megatonnes) of copper, based on current resources
plus past production (16 756 Mt at 0·558% Cu;
CODELCO, 2010). It belongs to the north^south-trending
Neogene metallogenic belt of Central Chile, which extends
along the western slope of the Chilean Andes (32^34·58S,
Fig. 1). This belt includes other giant ore deposits and constitutes one of the most richly endowed copper provinces
in the world, with more than 220 Mt of contained Cu
(Camus, 2003; Antofagasta plc, 2009; CODELCO, 2010).
The objective of the present study is to track the petrogenetic evolution of El Teniente related magmas, from the
source to emplacement levels, as recorded by zircon
1092
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Fig. 1. (a) Schematic map showing the main tectonic features of the southeastern Nazca Plate and Chilean continental margin. The map shows
the distribution of volcanoes constituting the Northern and Southern Chilean volcanic zones (CVZ, SVZ), the volcanic gap between 288 and
338S, the location of porphyry Cu^Mo deposits of the Central Chile Neogene Metallogenic Belt (black stars), and the Wadati^Benioff zone
contours of the convergent zone (dotted line, Isacks, 1988). LP, Los Pelambres; RB-LB, R|¤o Blanco^Los Bronces; ET, El Teniente; RR, Rosario
de Rengo. (b) Geological map showing the distribution of the main lithological units and structural features of the central Chilean^
Argentinean Andes where El Teniente and other porphyry Cu^Mo deposits are emplaced. Upper inset shows the distribution of main morphostructural units forming the Andean orogen in this area. Circled numbers indicate the location of the areas discussed in the text: 1, Abanico
and Farellones formations NE of the R|¤o Blanco^Los Bronces deposit; 2, San Francisco batholith and Yerba Loca pluton; 3, La Gloria pluton
and Cerro Meso¤n Alto stock. PFZ, Pocuro Fault Zone; AFTB, Aconcagua Fold and Thrust Belt.
1093
JOURNAL OF PETROLOGY
VOLUME 53
grains and to frame this evolution within the continental
margin global geodynamic setting during ore deposit formation. For this purpose, isotopic (O, Hf) and trace element [Ti, Y, Hf, rare earth elements (REE)] compositions
have been determined in zircons with previously known
U^Pb ages and U^Th contents (Maksaev et al., 2004).
These analyses are complemented by a study of the external morphology and internal structure of the zircons by
standard optical methods and cathodoluminescence (CL)
images. The main results reveal patterns of magmatic evolution and a common source for the different intrusive
pulses of the El Teniente deposit. This source is indistinguishable from that of the preceding barren magmatic activity in the region. We argue that this is a consequence of
long-lived MASH-type processes, as originally defined by
Hildreth & Moorbath (1988), where ascending, subduction-related, mantle-derived magmas are hybridized in
deep lithospheric zones of melting, assimilation, storage
and homogenization. Such processes can control the isotopic characteristics of the magmatism during extended
periods of time and throughout the contrasting tectonic
regimes within the margin, as observed for this portion
of the Andean range. We argue that El Teniente mineralization-related fertile magmas are a mixture of melt components derived from dehydration melting in the enriched
MASH reservoir and from primary basalt differentiation.
Additionally, we show that this process can fractionate O
to generate low-d18O magmas, and finally we discuss how
dehydration melting at the base of a thickened crust favors
the formation of giant porphyry copper deposits simultaneously and in a regional context.
G E N E R A L B AC KG RO U N D
Regional geology and geodynamic setting
The three major known porphyry Cu^Mo deposits of
the Central Chile Neogene Metallogenic Belt are located
between 328 and 348S: Los Pelambres^El Pacho¤n (328S),
R|¤o Blanco^Los Bronces (33808’S), and El Teniente
(34804’S; Fig. 1a). These deposits formed between late
Miocene and Pliocene times, the northernmost one being
older (10^12 Ma; Perello¤ et al., 2009) than the remaining
two, which are considered coeval (6^4 Ma; Maksaev
et al., 2004; Deckart et al., 2005). They are distributed
along the western slope of the Andean range through two
morphostructurally different segments of the continental
margin separated at 338S. This latitude also coincides
with the current locus of subduction of the Juan
Ferna¤ndez Ridge, the southern limit of the flat-slab subduction segment (27^338S) and the beginning of the
Chilean Southern Volcanic Zone (33^468S; Fig. 1a).
The western slope of the Andean Principal Cordillera in
Central Chile is dominated by Cenozoic igneous rocks distributed along an 60 km wide north^south-trending belt
(Fig. 1b). Older units are exposed to the east, Mesozoic
NUMBER 6
JUNE 2012
sedimentary sequences crop out near the Chilean^
Argentinean border and Triassic volcanic rocks and crystalline Paleozoic basement compose the Argentinean
Frontal Cordillera (Fig. 1b). In marked contrast to other
metallogenic belts in the Chilean continental margin,
major trench-parallel structures spatially related to the
Neogene Metallogenic Belt appear to be absent. The main
structural systems developed in this part of the Andean
range are the west-vergent Pocuro^San Ramo¤n Fault,
bounding the Principal Cordillera to the west, and the
east-vergent Aconcagua Fold and Thrust Belt affecting the
Mesozoic deposits near the Chilean^Argentinean border
(Fig. 1b). However, local structures have been described
for each deposit and have been related to reactivation of
pre-Cenozoic basement structures during the development
of the Andean Cordillera (e.g. Rivera & Falco¤n, 2000).
The Central Chilean Andes Cenozoic magmatic rocks
are the product of a prolonged and intense period of
arc-related igneous activity lasting from 536 to 6 Ma.
They form a nearly 5500 m thick volcanic^volcaniclastic
sequence that makes up the Farellones and Abanico
(¼ Coya Machal|¤ ) formations (Charrier et al., 2002). The
Abanico Formation was deposited during Oligocene^early
Miocene times in a tholeiitic arc setting, overlying an
30^35 km thick continental crust, during basin development under crustal extension (Charrier et al., 2002; Kay
et al., 2005). Continued volcanic activity after basin inversion, between 21 and 15 Ma (Charrier et al., 2002), led to deposition of the Farellones Formation during the Miocene.
Igneous rocks formed during this time show a progressively more calc-alkaline affinity with respect to the preceding magmatism and were formed in an arc setting
over a progressively thickening crust. This has been
inferred to have reached no more than 45^50 km
(Charrier et al., 2002; Kay et al., 2005), the current estimated crustal thickness under the Central Chilean
Cenozoic magmatic belt (Tassara et al., 2006). Diminished
magmatic activity followed these episodes and is represented in numerous isolated intrusive bodies and less abundant volcanic rocks throughout the region. Overall, these
rocks young to the east until reaching the current active
volcanic zone near the Chilean^Argentinean border, revealing the progressive arc migration that followed
Farellones Formation deposition (Stern & Skewes, 1995;
Kay et al., 2005).
The Abanico basin inversion during early to middle
Miocene times marks the onset of the constructive period
of the Andean orogen in Central Chile, characterized by
shortening, thickening, and uplifting processes (Mpodozis
& Ramos, 1989; Giambiagi & Ramos, 2002; Kay et al.,
2005; Far|¤ as et al., 2008, 2010). During this period, shortening was accommodated by different structural systems
and migrated to the east in three stages: (1) early to
middle Miocene inversion of basin-bounding normal
1094
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
faults currently bounding Abanico Formation outcrops; (2)
middle to late Miocene development of the Aconcagua
Fold and Thrust Belt; (3) late Miocene to Pliocene activity
of high-angle reverse faults that uplifted the Frontal
Cordillera in Argentina and a series of out-of-sequence
thrusting in the eastern Principal Cordillera (Fig. 1b;
Mpodozis & Ramos, 1989; Charrier et al., 2002;
Giambiagi & Ramos, 2002; Fock et al., 2006).
Concomitant progressive crustal thickening during this
evolution has been inferred mostly from the geochemical
signatures of coeval igneous rocks from the Abanico and
Farellones formations (Kay & Mpodozis, 2002; Kay et al.,
2005). Overall uplift of the Andean Principal Cordillera
occurred during late Miocene^Pliocene times, resulting in
an accumulated 2 km of uplift in 2 Myr during an event
taking place sometime between 10 and 4 Ma (Far|¤ as et al.,
2008). Additionally, high exhumation rates have been
recognized locally and regionally in the area during this
same time interval, particularly in the Western Principal
Cordillera south of 338S where the R|¤o Blanco^Los
Bronces and El Teniente deposits are located (Skewes &
Holmgren, 1993; Maksaev et al., 2009).
El Teniente porphyry Cu^Mo deposit
El Teniente is genetically linked to late Miocene^early
Pliocene magmatic^hydrothermal processes (Howell &
Molloy, 1960; Cuadra, 1986; Skewes et al., 2002; Camus,
2003). The deposit is hosted by a mafic volcano-plutonic
complex, known as the Teniente Mafic Complex, composed of pervasively altered dark grey to black basalt, andesite, diabase sills, and gabbro intrusions forming a
450 km3 laccolith emplaced in the Farellones Formation
(Fig. 2; Lindgren & Bastin, 1922; Skewes et al., 2002).
Intensive alteration has prevented an accurate age determination for this intrusive unit. However, an apatite
fission-track age of 8·9 2·8 Ma has been obtained for corresponding rocks near the mine (all ages are indicated
with 2s error level; Skewes et al., 2002; Maksaev et al.,
2004), which agrees with K^Ar ages ranging between 12
and 6 Ma for the Farellones Formation in the region
(Cuadra, 1986; Kay et al., 2005). A series of felsic to intermediate stocks and dikes intruded the Teniente Mafic
Complex between 6·5 and 2·9 Ma (Fig. 2; Cuadra, 1986;
Maksaev et al., 2004). They are quartz-diorites, tonalites
and granodiorites, with subordinate diorites and
hornblende-rich andesitic dykes. The main intrusive igneous events are represented by intrusions forming the
Sewell Stock, A Porphyry, Central and Northern diorites,
Teniente Porphyry, and Late Dacite and Hornblende
dikes (these correspond to informal names widely
adopted in the literature; Fig. 2). The youngest igneous
activity within the deposit is represented by the postmineralization and alteration-free Late Hornblende Dikes
(3·8^2·9 Ma: Cuadra, 1986; Maksaev et al., 2004). Multiple
magmatic and hydrothermal breccia complexes complete
Fig. 2. Geology of the El Teniente 4 LHD level (2354 m) from the El
Teniente mine. Locations of the samples discussed from main intrusive units are indicated with white diamonds along with their corresponding labels and zircon U^Pb age (2s; Maksaev et al., 2004).
Two ages are shown for samples with a bimodal distribution; the
older ages correspond to the dominant age peak.
the deposit geology (Fig. 2). The Braden Pipe, the largest
breccia and main lithological feature, is an inverted
cone-shaped, weakly mineralized, diatreme body composed of two facies (Fig. 2). These represent late to
post-mineralization events dated between 4·4 and 4·8 Ma
(Cuadra, 1986; Maksaev et al., 2004).
Hypogene mineralization at El Teniente is mostly distributed within a dense vein stockwork and a variety of
magmatic^hydrothermal breccias (Camus, 1975; Cuadra,
1986; Skewes et al., 2002). Alteration assemblages have
been classically divided into three main hypogene stages;
however, this is an oversimplified scheme as a variety of
localized hydrothermal events have been identified, reflecting an evolution by multi-stage processes (Cuadra, 1986;
Skewes et al., 2002, 2005; Maksaev et al., 2004; Cannell
et al., 2005; Vry et al., 2010). The superimposition of multiple, discrete, magmatic and hydrothermal events has led
1095
JOURNAL OF PETROLOGY
VOLUME 53
to complex alteration and mineralization patterns; however, this is also probably responsible for the high metal
concentrations in El Teniente (Skewes et al., 2002, 2005;
Maksaev et al., 2004; Cannell et al., 2005; Vry et al., 2010).
Isotopic evidence has shown that the main budget of
metals, water and sulfur is of magmatic origin (Kusakabe
et al., 1984, 1990; Skewes et al., 2001). However, in the last
decade, there has been an intensive debate over the agents
responsible for their introduction into the deposit (Skewes
et al., 2002, 2005; Maksaev et al., 2004; Cannell et al., 2005,
2007; Stern & Skewes, 2005; Skewes & Stern, 2007; Vry
et al., 2010). Since the early investigations, El Teniente has
been considered to be a typical porphyry deposit in terms
of its alteration assemblages, vein and breccia style, and
spatial and temporal relationships between Cu^Mo mineralization and felsic porphyries (e.g. Howell & Molloy,
1960; Camus, 1975; Cuadra, 1986). The felsic porphyries
are considered to have been conduits for mineralizing
fluids sourced from a deeper-level magma chamber (e.g.
Cannell et al., 2005; Klemm et al., 2007; Vry et al., 2010).
Skewes et al. (2002) argued that the felsic intrusions correspond to small, late, copper-poor stocks that merely redistributed earlier copper mineralization that had already
been introduced by previously unmapped early formed
breccia pipes. This mineralization would ultimately originate as a fluid discharge from an unexposed evolving
magma chamber of batholithic dimensions (Skewes et al.,
2005; Stern & Skewes, 2005; Stern et al., 2010). The differing
interpretations arise mainly as a consequence of the complexity of the mineralization and alteration patterns, together with the variable interpretations of the time
relationships for the intrusive events as determined by the
numerous K^Ar, Ar^Ar, Re^Os, and U^Pb ages reported.
However, these studies agree that ultimately the breccias,
veining, and intrusions are linked to a deep magma chamber located below the mine level and are thus derived
from the evolution of a common magmatic system.
Recently published data on alteration and mineralization
patterns and their relation to intrusive events have been
used to argue strongly in favor of the porphyry-style
model, indicating that El Teniente represents a nested,
but otherwise typical, porphyry Cu^Mo deposit (Vry
et al., 2010).
Sampled units and previous zircon U^Pb
age data
We address the petrogenesis of the intrusive rocks related
to the El Teniente deposit through new morphological,
chemical and isotopic studies of selected single zircon
grains that have previously been dated [the complete
U^Pb dataset has been given by Maksaev et al. (2004); a selection of these data relevant for this study is presented in
Table 1]. The zircon grains are from six intrusive units
that cover a timespan of 1·6 Myr and are located inside
the mine, within the limits of the ore body, that is the
NUMBER 6
JUNE 2012
Sewell Stock, the A Porphyry, the Northern and Central
diorites, the Teniente Porphyry, and a Late Dacite Dike
(Fig. 2). The Sewell Stock, located in the southeastern part
of the deposit (Fig. 2), is the oldest intrusion and largest
(30 km3) compared with the younger intrusive bodies
(51km3). It shows two textural varieties with transitional
contacts, suggesting emplacement as a composite intrusion
(Faunes, 1981). The A Porphyry and the Central and
Northern diorites are thin, cylindrical to irregularly
shaped intrusions located in the southeastern and eastern
portion of the deposit (Fig. 2). The Teniente Porphyry is a
north^south-trending tabular stock whose southern edge
has been truncated by the Braden Breccia (Fig. 2). Similar
to the Sewell Stock, two textural varieties have been identified in the Teniente Porphyry and have been attributed to
at least two independent intrusive pulses from a common
magmatic source (Rojas, 2003). The Late Dacite Dike belongs to a series of 2^15 m wide felsic dikes located mainly
in the SW and NE part of the deposit. They occur as concentric dikes surrounding the Braden Breccia and also as
NE^SW- to NW^SE-trending planar dikes (Fig. 2).
Compositionally, all these intrusive units are felsic dacitic
igneous rocks ranging between 60 and 69 wt % SiO2,
except for the relatively more mafic A Porphyry, which
has an andesitic composition with SiO2 contents between
56 and 62 wt % (Rojas, 2003; Cannell et al., 2005;
Gonza¤lez, 2006; Hitschfeld, 2006; Stern et al., 2007, 2010).
They show strongly fractionated REE patterns with La/
YbN 9^44 along with high Sr/Y 24^253. These ‘adakite-like’ characteristics are shared by a few coeval intrusive
rocks exposed near the deposit (Rabbia et al., 2000; Reich,
2001), but are otherwise absent in any of the earlier or
later igneous rocks in the region (e.g. Kay et al., 2005). As
with Central Chile Cenozoic magmatism in general, intrusive rocks from El Teniente are considered primarily as
being derived from subduction-related, mantle-derived
magmas. ‘Adakite-like’ characteristics have been attributed
by Kay et al. (2005) to a combination of source contamination by subduction erosion and incorporation of a component derived from melting at the base of the thickened
lower crust. Alternatively, Stern & Skewes (2005) and
Stern et al. (2010) argued that they result mainly from the
fractionation of igneous phases and extensive fluid transfer
to the top of a crystallizing batholithic-size magma body.
Zircon U^Pb age determinations have constrained crystallization ages that were previously biased by the use of
K^Ar and Ar^Ar chronometers in the pervasively altered
rocks of El Teniente (Fig. 2; Maksaev et al., 2004).
However, the final interpretation of these ages has been
open to debate (Cannell et al., 2005, 2007; Skewes et al.,
2005, 2007). Maksaev et al. (2004) showed that zircon U^
Pb ages for the Sewell Stock, A Porphyry and Northern
and Central diorites have a bimodal distribution with
peaks at 6·4^6·1Ma and 5·6^5·4 Ma for the dominant and
1096
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Table 1: Sample locations and U,Th,Th/U and U^Pb age data for the zircon spots from the El Teniente deposit analyzed
in this study (data from Maksaev et al., 2004)
Spot
Crystal
U
Th
sector
(ppm)
(ppm)
Th/U
Age
2s
Spot
(Ma)
A Porphyry, Sample TT150
Crystal
U
Th
sector
(ppm)
(ppm)
Th/U
Age
2s
(Ma)
Northern Diorite, Sample TT102
1.1
r
62
48
0·78
6·6
0·8
1.1
r
80
37
0·46
6·4
0·6
3.1
r
76
62
0·82
6·8
0·6
2.1
c
55
37
0·68
6·7
0·8
4.1
r
73
60
0·82
6·4
0·6
3.1
r
174
133
0·76
5·9
0·6
5.2
r
64
52
0·81
5·7
0·8
4.1
r
95
51
0·53
5·9
0·6
6.1
r
542
920
1·70
6·5
0·2
6.1
c
185
148
0·80
9·0
0·8
7.1
r
53
32
0·60
6·1
0·8
8.1
r
138
86
0·62
5·9
0·6
11.1
c
36
18
0·51
6·7
1·0
8.2
c
511
235
0·46
27·8
0·6
12.1
r
46
33
0·73
6·8
1·0
11.1
c
381
252
0·66
79·4
1·4
13.1*
r
1192
459
0·39
6·4
0·2
11.2
r
1569
564
0·36
6·1
0·2
13.2
c
69
58
0·84
6·2
0·6
12.2
c
60
34
0·56
5·3
0·8
13.1
c
566
209
0·37
6·3
0·4
Sewell Stock, Sample TT101
Teniente Porphyry, Sample TT94
1.1
r
3616
2782
0·77
6·4
0·4
1.1
r
154
107
0·70
5·4
0·4
1.2
c
45
24
0·55
6·0
0·8
2.1
r
146
117
0·80
5·3
0·4
3.1
r
89
41
0·46
5·8
0·6
4.1
r
97
56
0·57
4·7
0·6
3.2
c
68
54
0·79
6·3
0·8
5.1
r
197
93
0·47
4·8
0·4
4.2
c
53
42
0·78
6·3
0·8
6.1
r
142
86
0·61
5·6
0·4
7.1*
r
2587
1101
0·43
6·1
0·2
6.2
c
115
87
0·75
5·0
0·8
7.2
c
93
71
0·77
5·6
0·6
7.1
r
239
212
0·88
5·4
0·4
9.1*
r
3848
2519
0·65
6·0
0·3
8.1
c
255
318
1·25
5·1
0·4
9.2
c
155
158
1·02
6·3
0·4
9.1
c
262
241
0·92
6·0
0·4
10.1
r
362
298
0·82
6·2
0·4
9.2
r
300
183
0·61
5·4
0·4
12.1
r
86
34
0·40
5·2
1·0
15.1
r
178
104
0·59
5·2
0·4
17.1
r
206
182
0·88
5·3
0·4
Central Diorite, Sample TT90
Late Dacite Dike, Sample TT91
3.1
c
85
45
0·53
6·4
1·0
2.1
r
362
275
0·76
4·5
0·2
3.2
r
57
34
0·61
4·8
1·0
3.1
r
485
434
0·89
5·2
0·4
5.1
c
85
78
0·91
5·1
0·8
4.1
r
227
160
0·70
4·7
0·4
5.2
r
71
45
0·63
5·6
0·6
4.2
c
77
42
0·54
6·2
0·6
7.1
r
88
30
0·34
7·0
0·6
5.1
r
391
264
0·68
5·0
0·2
8.1
r
62
50
0·82
5·7
0·6
7.1
r
274
172
0·63
4·4
0·2
11.1
c
54
33
0·61
6·8
0·8
8.1
r
295
180
0·61
4·7
0·2
11.2
r
45
22
0·50
6·0
0·8
10.1
r
356
280
0·79
4·7
0·2
12.1
r
98
62
0·63
5·7
0·6
12.1
r
272
203
0·75
4·9
0·4
14.1
r
52
42
0·81
6·5
0·8
15.1
r
263
177
0·67
4·8
0·4
17.1
r
43
26
0·61
6·5
1·4
17.1
r
215
132
0·61
4·8
0·4
Sample TT150: zircon U–Pb age with bimodal distribution, 6·46 0·11 Ma and 5·67 0·19 Ma for the dominant and
subordinate age groups, respectively. DDH-1337, 384 m; sample level 2044 m; northing 110N/easting 1735E. Sample
TT102: zircon U–Pb age with bimodal distribution, 6·11 0·13 Ma and 5·49 0·19 Ma for the dominant and subordinate
age groups, respectively. Level Teniente 6 UCL, 2161 m; northing 1016N/easting 1110 E. Sample TT101: zircon U–Pb age
with bimodal distribution, 6·15 0·08 Ma and 5·59 0·17 Ma for the dominant and subordinate age groups, respectively.
Level Teniente 4, 2347 m; northing –265N/easting 1365E. Sample TT94: zircon U–Pb age with unimodal distribution,
5·28 0·10 Ma. Level Teniente 6, 2161 m; northing 1050N/easting 450E. Sample TT90: zircon U–Pb age with bimodal
distribution, 6·28 0·16 Ma and 5·50 0·24 Ma for the dominant and subordinate age groups, respectively. Level UCL
Esmeralda, 2192 m; northing 250N/easting 1325E. Sample TT91: zircon U–Pb age with unimodal distribution,
4·82 0·09 Ma. Level UCL Esmeralda, 2192 m; northing 310N/easting 1030E. c, core; r, rim.
*Weakly luminescent overgrowth rims observed in CL images.
1097
JOURNAL OF PETROLOGY
VOLUME 53
subordinate age populations, respectively. Older ages have
been interpreted to correspond to the intrusion age and
younger ones to be related to hydrothermal activity.
The remaining units show unimodal age distributions
with peaks at 5·28 0·19 Ma for the Teniente Porphyry
and 4·82 0·09 Ma for the Late Dacite Dike.
A N A LY T I C A L T E C H N I Q U E S
Zircon separation was carried out at the University of
Chile mineral separation facility, and further selection
along with imaging studies and compositional analyses
were performed at the Research School of Earth Sciences
of the National Australian University (RSES-ANU).
Zircon grains were separated from total rock samples
using standard crushing, washing, heavy liquid (specific
gravity 2·96 and 3·3), and paramagnetic procedures.
Hand selected zircon grains were placed onto double-sided
tape, mounted in epoxy together with chips of the reference zircons (Temora and SL13), sectioned approximately
in half, and polished. Reflected and transmitted light
photomicrographs were prepared for all zircons, as were
CL scanning electron microscope (SEM) images. These
CL images were used to decipher the internal structures
of the sectioned grains and to ensure that the 20 mm sensitive high-resolution ion microprobe (SHRIMP) spot was
wholly within a single age component within the sectioned
grains.
REE data were acquired using SHRIMP II in spots adjacent to those analyzed for U^Pb^Th geochronology
and belonging to the same crystal sector according to the
CL images. The energy filtering method was used to
reduce isobaric interferences (Ireland & Wlotzka, 1992).
Operating conditions and data reduction methods are
similar to those described by Hoskin (1998). REE detection
limits are in the vicinity of 0·01ppm for analysis spots that
are 30 mm across and a few micrometers deep. The in situ
Ti analyses were also made using SHRIMP II in a separate session using methods similar to those described by
Hiess et al. (2008). Previous U^Pb^Th analytical spots
were lightly polished, then the same area within the
grains was analyzed.
Oxygen isotope analyses were made using the SHRIMP
II fitted with a Cs source and electron gun for charge compensation following methods described by Ickert et al.
(2008). The SHRIMP U^Pb, REE and Ti analytical spots,
craters approximately 20 mm in diameter by 1^2 mm deep,
were polished from the mount surface. The oxygen isotope
analyses were then made on exactly the same location as
used for the U^Pb analyses. Oxygen isotope ratios were
determined in multiple collector mode using an axial continuous electron multiplier (CEM) triplet collector, and
two floating heads with interchangeable CEM^Faraday
cups. The Temora II reference zircon was analyzed to
monitor and correct for isotope fractionation. The
NUMBER 6
JUNE 2012
measured 18O/16O ratios and calculated d18O values have
been normalized relative to a Temora II weighted mean
d18O value of þ8·2ø (Ickert et al., 2008). Reproducibility
for the Temora II reference zircon d18O value was 0·551
and 0·715ø (2s uncertainty) for the analytical sessions.
As a secondary reference, zircons from the Duluth
Gabbro sample FC1 analyzed in the same analytical sessions gave a d18O value of 5·405 0·348ø and
5·415 0·615ø (2s uncertainty), in agreement with data
reported by Ickert et al. (2008).
Lu^Hf isotopic measurements were conducted by laser
ablation multi-collector inductively coupled plasma mass
spectroscopy (LA-MC-ICPMS) using the RSES Neptune
MC-ICPMS coupled with a 193 nm ArF Excimer laser;
similar to procedures described by Munizaga et al. (2008).
Laser ablation analyses were performed on the same locations within single zircon grains as used for both the U^
Pb and oxygen isotope analyses. For all analyses of unknowns or secondary standards, the laser spot size was c.
47 mm in diameter. The mass spectrometer was first tuned
to optimal sensitivity using a large grain of zircon from
the Monastery kimberlite. Isotopic masses were measured
simultaneously in static-collection mode. A gas blank was
acquired at regular intervals throughout the analytical session (every 10 analyses). The laser was fired with typically 5^8 Hz repetition rate and 60 mJ energy. Data were
acquired for 100 s, but in many cases only a selected interval from the total acquisition was used in data reduction.
Throughout the analytical session mostly FC1 and other
widely used reference zircons (91500, Temora-2, Monastery
and Mud Tank) were analyzed to monitor data quality.
FC1 gave a weighted mean 176Hf/177Hf fractionation corrected ratio of 0·282173 12 (2s uncertainty) for 11 analyses, which is within uncertainty of reported solution
values (Woodhead & Hergt, 2005; Vervoort, 2010). Signal
intensity was typically c. 5^6 V for total Hf at the beginning of ablation, and decreased over the acquisition time
to 2 V or less. Isobaric interferences of 176Lu and 176Yb on
the 176Hf signal were corrected by monitoring signal intensities of 175Lu and 173Yb, 172Yb and 171Yb. The calculation
of signal intensity for 176Hf also involved independent
mass bias corrections for Lu and Yb.
R E S U LT S
Zircon crystal morphology and internal
structure
Zircon grains from the six studied units have mainly euhedral external morphologies and zoned inner structures
typical of igneous zircons (Fig. 3). Overall, the crystals
show the development of prism ({100} and {110}) and pyramidal faces ({211} and {101}), and internal oscillatory
and sector zoning. However, there are marked variations
of these characteristics between units. Disruptions of the
1098
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Fig. 3. Cathodoluminescence and corresponding transmitted light images of zircon grains from the El Teniente deposit. Intrusive units correspond to: (a) A Porphyry; (b) Sewell Stock; (c) Central Diorite; (d) Northern Diorite; (e) Teniente Porphyry; (f) Late Dacite Dike. Illustrated
zircons are representative of the main morphological variations within each unit. Circles correspond to the U^Pb age spot location and analysis
labels are indicated in the transmitted light images. (1) zircons with high U^Th, weakly luminescent rims, (2) inherited cores. A clear mottled
texture is developed on the surfaces of grains (a) 11; (b) 4, 9, 3, 7, 1, 29; (c) 8, 1, 5; (d) 5, 6, 8.
inner structure, such as internal cracks or resorption textures, are absent and inherited cores are extremely scarce
(Table 1). Only a few grains show features suggesting a
hydrothermal imprint (Fig. 3a and d). All these characteristics are detailed in the following discussion, where the intrusive units have been grouped according to the common
morphological and structural parameters of their zircon
populations.
A Porphyry
Zircons of the A Porphyry are the largest among the El
Teniente zircons. They are commonly c. 100^300 mm equidimensional to stubby euhedral crystals, with aspect ratios
mostly between 1 and 1·5 (Figs 3a and 4). CL images show
the development of weak euhedral oscillatory zoning overprinted by strong sector zoning (Fig. 3a). Their morphology is characterized by well-developed {100} and {110}
1099
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 6
JUNE 2012
prism forms and the presence of two pyramids, with
the {101} form being predominant over the {211} form
(Figs 3a and 4). Subordinately, crystals show mottled surfaces, observed in transmitted light images, that correspond to irregular crystal rims of low luminescent contrast
in CL images (Fig. 3a: grain 11). These features are associated with high U and Th contents, but not with any
other particular chemical or isotopic composition, or
U^Pb age, among those determined in the whole population of A Porphyry zircon grains.
Sewell Stock, Central Diorite and Northern Diorite
These units have similar zircon populations that are less
homogeneous in morphological types and size distribution
than zircons from the rest of the units. Crystals are euhedral to subhedral with short prismatic and subordinately
equidimensional habits and show oscillatory and sector
zoning (Fig. 3b^d). They are of 100^250 mm length and
100^150 mm width, with aspect ratios between 1 and 3·5
(Fig. 4). Morphological types are varied, although overall
there is a similar development of both prismatic and pyramidal forms (Figs 3b^d and 4). Several grains show euhedral to rounded cores that preserve inner, zoned, igneous
structures. Their rims are oscillatory zoned euhedral overgrowths, which are usually brighter under CL (Fig. 3b:
grain 3; Fig. 3c: grains 3 and 5; Fig. 3d: grains 5, 6 and 8).
Chemical, isotopic and age determinations in both of
these crystal sectors show no discernible difference, except
in the Northern Diorite where three inherited cores are
identified by their significantly older ages (Fig. 3d: grain
8). As in the A Porphyry, some grains from the Sewell
Stock show a mottled texture on the surfaces (Fig. 3b: all
grains), observed as weakly luminescent rims in CL
images, which are associated with high U^Th contents.
Teniente Porphyry and Late Dacite Dike
Zircons from these units are euhedral, elongated, prismatic crystals of 100^250 mm length and 80^100 mm
width (Fig. 3e and f). Aspect ratios are mostly in the
range of 1·5^2·5 and 2^3 for the Teniente Porphyry
and the Late Dacite dike, respectively (Fig. 4). Zircon
grains are relatively simple with a single continuous
pattern of oscillatory and/or sector zoning throughout
the crystal (Fig. 3e and f). Morphological types are
characterized by the {110} prismatic form and the predominance of the pyramidal form {211} over the
{101} (Fig. 4). Few crystals develop the {100} prism,
which is more common in zircons from the Teniente
Porphyry than those from the Late Dacite Dike (Figs
3e, f and 4).
Zircon chemistry and Ti-in-zircon
thermometry
Trace element data for El Teniente zircons are reported in
Table 2. They include the determinations of REE, Y and
Fig. 4. (a) Main morphological types of El Teniente zircons schematically illustrated in the typological classification diagram of
Pupin (1980). (b) Histograms of aspect ratios (length/width) comparing distributions among the El Teniente intrusive units. n, number of
crystals measured.
Hf from this study. All zircons define a single group in
terms of their REE contents and normalized REE patterns. REE abundances are moderately low among the
values reported for crustal zircons (250^5000 ppm;
1100
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Table 2: REE, Yand Hf concentrations for zircons from El Teniente deposit
Spot
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
EuN/
CeN/
CeIV/
EuN*
CeN*
CeIII
(a)
(b)
(c)
Y
Hf
A Porphyry
1.1
0·074
7·9
0·347
3·47
3·63
1·44
12·9
3·1
34·4
12·9
53·5
13·1
111·8
26·9
0·64
12·0
69·7
361·0
3.1
0·074
14·0
0·288
3·50
5·06
1·41
22·6
6·2
68·0
24·6
98·2
22·3
166·2
36·7
0·40
23·2
82·5
652·1
2428·3
2596·6
5.2
0·066
8·8
0·413
3·74
3·79
1·43
14·0
3·5
38·3
14·3
58·1
14·2
119·8
28·0
0·60
13·0
74·0
389·6
2466·0
6.1
0·136
47·6
1·585
14·27
23·87
9·60
117·0
29·8
308·3
108·9
422·8
97·5
714·7
167·9
0·55
24·8
57·5
3002·3
2209·0
7.1
0·059
5·7
0·043
0·44
0·69
0·30
3·3
1·0
11·4
4·5
20·0
5·3
49·9
11·3
0·60
27·4
576·4
130·7
2364·6
13.2
0·072
8·6
0·408
4·02
4·18
1·38
14·4
3·6
39·2
14·6
62·4
15·3
125·8
30·7
0·54
12·2
65·5
411·1
2307·6
Sewell Stock
3.2
0·062
11·2
0·229
2·69
3·62
1·33
14·2
3·7
39·8
14·6
61·1
15·1
126·6
31·3
0·57
22·7
142·4
405·6
2754·2
4.2
0·060
7·9
0·166
1·91
2·28
0·86
8·8
2·3
24·6
9·5
41·6
10·6
92·4
21·8
0·59
19·3
111·6
261·8
2676·6
7.1*
0·097
25·3
0·202
1·92
3·10
1·18
16·0
4·7
55·0
22·3
97·9
23·7
203·3
49·7
0·51
43·6
26·8
709·0
3365·7
7.2
0·075
9·3
0·109
1·12
1·59
0·66
7·8
2·4
30·8
13·7
63·7
17·0
146·8
38·1
0·57
25·0
228·3
395·8
2351·3
9.1*
0·086
6·2
0·175
1·76
2·01
0·76
7·2
2·0
21·6
8·4
37·5
9·7
84·8
20·6
0·61
12·3
147·1
234·8
2525·5
9.2
0·118
14·4
0·794
6·26
6·20
2·44
25·8
7·2
81·5
31·1
131·5
32·1
256·5
63·5
0·59
11·4
65·7
899·9
2217·1
10.1
0·099
17·5
0·053
0·53
1·04
0·41
6·4
2·0
25·1
10·8
49·0
13·1
117·3
28·8
0·49
58·8
223·7
313·7
3102·1
2629·2
Central Diorite
3.1
0·080
8·4
0·058
0·67
1·05
0·41
4·8
1·5
17·3
7·2
32·8
8·7
79·9
19·9
0·55
30·0
366·6
209·0
3.2
0·050
13·1
0·057
0·48
0·87
0·26
4·8
1·5
18·2
7·1
31·4
7·7
66·5
14·9
0·39
59·4
596·3
195·2
2796·8
5.2
0·079
8·8
0·339
3·52
4·03
1·44
14·2
3·5
37·7
13·7
55·6
13·2
110·2
26·7
0·58
13·0
35·0
368·6
2482·4
11.2
0·064
10·3
0·052
0·55
0·91
0·35
5·0
1·5
18·4
7·3
33·1
8·2
72·9
16·2
0·50
43·0
466·6
202·8
2771·8
14.1
0·081
9·1
0·342
3·60
4·00
1·42
15·1
3·8
41·6
15·9
62·9
14·9
119·6
27·7
0·56
13·2
37·4
421·7
2349·1
17.1
0·096
8·4
0·129
1·11
1·44
0·52
5·6
1·7
18·1
7·2
32·8
8·3
76·2
19·0
0·55
18·2
188·7
199·3
2290·6
Northern Diorite
1.1
0·067
9·2
0·042
0·49
0·70
0·27
3·6
1·1
13·0
5·5
24·2
6·2
61·3
13·8
0·52
41·9
619·3
151·0
2903·3
2.1
0·075
7·8
0·140
1·63
2·33
0·97
9·5
2·5
26·5
9·9
41·7
10·3
91·5
21·5
0·63
18·5
75·1
283·1
2625·8
8.1
0·061
9·4
0·097
1·05
1·50
0·61
7·3
2·2
27·0
11·9
54·8
14·5
126·6
32·6
0·56
29·5
330·6
340·0
2449·7
8.2
0·071
8·9
0·108
1·20
1·89
0·27
11·7
4·0
52·7
22·5
104·0
26·0
207·3
48·7
0·17
24·5
275·8
622·7
2775·5
11.1
0·516
12·6
0·414
1·72
2·07
0·52
12·5
4·5
63·7
29·7
145·7
38·9
328·2
85·9
0·31
6·6
588·2
849·3
2408·6
12.2
0·072
11·0
0·040
0·41
0·70
0·32
4·4
1·4
17·2
7·4
34·9
9·3
87·4
21·5
0·55
49·6
1122·1
211·0
3097·6
Teniente Porphyry
1.1
0·062
9·2
0·062
0·59
1·00
0·38
5·1
1·5
17·4
7·5
33·0
8·5
77·5
18·6
0·52
36·0
412·4
209·9
2895·5
5.1
0·069
12·9
0·043
0·43
0·82
0·30
4·8
1·5
19·7
8·5
38·1
9·7
90·0
22·0
0·46
57·4
1002·5
252·0
3226·5
6.1
0·080
18·1
0·068
0·77
1·27
0·46
7·2
2·2
27·1
11·0
49·3
12·2
108·6
24·6
0·47
59·4
651·2
320·4
2857·5
7.1
0·075
21·9
0·108
1·10
2·04
0·72
11·3
3·6
45·1
19·1
88·5
22·2
186·7
46·3
0·46
59·0
574·1
547·5
2672·7
9.1
0·075
9·4
0·120
1·52
2·38
0·93
11·1
3·1
33·4
12·5
53·0
13·2
112·5
26·9
0·55
23·9
107·3
348·1
2860·7
12.1
0·073
9·2
0·027
0·23
0·40
0·17
2·5
0·8
10·7
5·0
25·6
7·5
78·7
20·0
0·52
50·3
2717·8
172·8
3174·0
3026·6
Late Dacite Dike
4.1
0·088
26·6
0·096
1·01
1·71
0·64
9·7
2·9
36·6
14·5
63·6
16·1
135·8
31·9
0·48
70·0
738·0
405·8
4.2
0·069
8·5
0·087
1·03
1·36
0·52
6·6
1·9
23·7
10·4
48·1
12·7
113·8
29·2
0·53
26·6
354·1
278·7
2786·0
5.1
0·066
8·8
0·413
3·74
3·79
1·43
14·0
3·5
38·3
14·3
58·1
14·2
119·8
28·0
0·60
13·0
45·3
389·6
2466·0
10.1
0·064
38·9
0·155
1·71
2·68
1·03
13·4
3·7
43·3
16·9
71·8
17·5
144·2
34·3
0·53
94·6
482·2
462·2
3110·4
12.1
0·087
51·0
0·140
1·48
3·07
1·12
19·7
6·3
77·9
32·7
143·6
35·0
283·4
68·1
0·44
111·6
924·7
922·9
3073·1
17.1
0·063
22·4
0·062
0·60
1·22
0·52
7·9
2·5
31·3
13·4
60·9
15·1
136·9
36·3
0·52
87·2
1390·7
375·3
3351·9
All concentrations are reported in ppm. (a) Eu-anomaly calculated as EuN/(SmNGdN)1/2; (b) Ce-anomaly calculated as
CeN/(LaNPrN)1/2; (c) CeIV/CeIII calculated according to Ballard et al. (2002) with whole-rock data for El Teniente intrusive
rocks taken from Rojas (2003), Cannell (2004), González (2006) and Hitschfeld (2006).
1101
JOURNAL OF PETROLOGY
VOLUME 53
Hoskin & Schaltegger, 2003) ranging from 114 to 727 ppm
(Table 2). Chondrite-normalized REE patterns are characterized by a steep increase from LaN to LuN with a strong
positive Ce-anomaly [(Ce/Ce*)N: 6^111] and a slight negative Eu-anomaly [(Eu/Eu*)N: 0·1^0·6; Fig. 5]. Zircon
Ce4þ/Ce3þ ratios, which primarily depend on the oxidation state of the magma (Ballard et al., 2002), show a wide
range of values between 26 and 2717 (Fig. 5). A slight decrease in (Eu/Eu*)N ratios and an increase in (Ce/Ce*)N
and Ce4þ/Ce3þ ratios is observed with decreasing age
(Fig. 5). Additionally, Ce4þ/Ce3þ and (Eu/Eu*)N values
are mostly within the range defined by zircons from
mineralization-related intrusions associated with porphyry
copper deposits in northern Chile (Ballard et al., 2002;
Fig. 5). Overall, zircon REE patterns and concentrations
are typical of those reported for crustal zircons in general.
U and Th contents can be grouped by crystal sectors
showing simple igneous structures (fine oscillatory or
sector zoning) and those from weakly luminescent overgrowth rims, both characteristics observed in CL images
(Table 1; Fig. 3). For the former, U and Th are in the range
of 36^1570 ppm and 18^920 ppm, respectively, and single
grains show inner compositional variations up to one
order of magnitude in concentration. Weakly luminescent
overgrowth rims, observed only in zircons from the A
Porphyry and the Sewell Stock, have comparatively
higher U and Th contents in the range of 1125^4160 ppm
and 329^3481ppm, respectively. These concentrations are
one to two orders of magnitude higher than those of their
respective grain cores (Table 1). Zircon Th/U ratios are
relatively uniform between 0·5 and 1, irrespective of the
unit or crystal sector considered (Table 1; Fig. 6). Hf concentration varies between 2300 and 2400 ppm (Table 1;
Fig. 6) and Y between 130 and 920 ppm (Table 2; Fig. 6),
both within the range reported for crustal zircons
(Hoskin & Schaltegger, 2003). Overall, the Hf, U, and Th
contents of zircons from the Teniente Porphyry and the
Late Dacite Dike extend to progressively higher values
relative to the older units (Fig. 6).
Temperatures were estimated using the Ti-in-zircon
thermometer from the equations of Ferry & Watson
(2007) whose calibration assumes crystallization under
rutile- and quartz-saturated conditions (aTiO2 ¼1;
aSiO2 ¼1) at 10 kbar. Though present, rutile has not been
reported in the ore deposit as part of the magmatic mineral assemblage and is generally considered to have a
hydrothermal origin (Rabbia, 2002). Thus, an aTiO2 ¼ 0·6
has been used in the calculations in agreement with the
general presence of other Fe^Ti oxides (e.g. Ferry &
Watson, 2007; Fu et al., 2008). Calculated temperatures
vary between 6138 and 8138C for a total range inTi concentration between 1·2 and 11·8 ppm (n ¼ 62; n is the number
of analyses; Fig. 7; Table 3). Temperature decreases systematically with decreasing age from an average of
NUMBER 6
JUNE 2012
Fig. 5. (a) Chondrite-normalized REE patterns for zircons from the
El Teniente deposit; normalization values after McDonough & Sun
(1995). The field defined by all analyses is highlighted and single
grain patterns from the Sewell Stock and the A Porphyry are shown
as examples. (b) Single grain and average CeN =CeN and CeIV =CeIII
Zrc
ratios vs age and EuN =EuN ratios. For single grains, the corresponding U^Pb ages were used for plotting; for averages, in samples with bimodal U^Pb age distributions, the oldest and dominant age peak
was used.CeIV =CeIII
Zrc calculated according to Ballard et al. (2002)
with whole-rock chemical data for El Teniente intrusions taken from
Rojas (2003), Cannell (2004), Gonza¤lez (2006) and Hitschfeld (2006).
777 508C in zircons from the A Porphyry to 681 788C
in zircons from the Late Dacite Dike (dispersion from
averages given at 2s, Fig. 7). Pressure corrections to these
temperatures would lower these estimates by up to 508C,
but most of them would still record higher temperatures
than the highest estimated for the hydrothermal activity
within the orebody (500^6008C; Cannell, 2004), and are
within the general range of felsic to intermediate igneous
rock crystallization temperatures (Fu et al., 2008).
Zircon Hf and O isotopes
The range in Hf and O isotopic compositions of El
Teniente zircons is remarkably uniform between the
1102
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Fig. 6. Plots of U, Y and Hf concentration vs age and U^Th covariation of zircons from the El Teniente deposit. Zircon U (and Th) contents
show a trend of increasing values with decreasing age despite the scatter induced by the high U analyses. A similar trend is seen in Hf contents
whereas Y remains essentially constant. Th/U ratios are mostly between 0·5 and 1·0. Symbols as in Fig. 5. In the U^age and U^Th plots
dashed lines indicate minimum concentrations shown by the weakly luminescent high U^Th rims identified from CL images. These have
been omitted from the averages of the Sewell Stock and A Porphyry samples. The complete dataset of U^Th concentrations has been given by
Maksaev et al. (2004).
various intrusive units. Importantly, isotopic values show
no correlation with respect to spot location within a single
zircon grain (core or rim), any chemical parameter, the
U^Pb ages, or between each other (Tables 4 and 5).
Zircon Hf ratios are characterized by a high initial
176
Hf/177Hf that varies between 0·283010 and 0·282945
(n ¼ 57; Table 4). Corresponding initial eHf values range
from þ8·4 to þ6·1, with an average of þ7·4 1·2, and
intra-grain variation is less than 1 eHf unit (Fig. 8). These
characteristics, together with the general absence of inherited zircons or zircon cores, are consistent with a
common magmatic system for the various intrusions
within the deposit. Two-stage depleted mantle (DM)
model ages range from 480 to 630 Ma (T2DM ; Table 4).
O isotope compositions for all Teniente zircons define a
range of d18OZrc values between 5·6 and 3·6ø, with an
arithmetic average of 4·7 1·0ø (Fig. 9; Table 5). At a
Fig. 7. Variation of temperature obtained by Ti-in-zircon thermometry vs age for zircons from the El Teniente deposit. Crystallization
temperatures define a trend of decreasing values with decreasing age.
Symbols as in Fig. 5.
1103
JOURNAL OF PETROLOGY
VOLUME 53
Table 3: Ti concentration and T8estimates for zircons from
El Teniente deposit
Spot
Ti
2s
(ppm)
T8Zrc
Spot
(8C)
A Porphyry
Ti
2s
(ppm)
T8Zrc
(8C)
Northern Diorite
1.1
6·3
0·3
750
1.1
5·0
0·1
729
3.1
7·4
0·2
766
2.1
5·6
0·4
739
4.1
5·9
0·4
745
3.1
4·9
0·2
727
5.2
7·3
0·3
765
4.1
6·9
0·3
759
6.1
11·8
0·4
813
6.1
7·9
0·1
773
7.1
9·3
0·4
789
8.1
7·1
0·3
762
11.1
6·1
0·1
747
8.2
8·9
0·4
784
12.1
8·2
0·2
775
11.1
4·7
0·3
723
13.1*
5·0
0·2
730
12.2
3·9
1·0
707
13.2
9·4
0·3
790
13.1
1·3
0·1
617
Sewell Stock
6·7
0·5
756
1.1
2·7
0·2
677
1.2
7·9
0·3
772
4.1
3·2
0·2
690
3.1
3·9
0·2
708
5.1
1·7
0·1
640
3.2
4·9
0·1
727
6.1
2·5
0·1
669
4.2
6·0
0·1
746
6.2
8·1
0·3
774
7.1*
1·2
0·1
613
7.1
3·2
0·2
691
7.2
7·2
0·2
763
8.1
4·9
0·2
728
9.1*
3·4
0·4
695
9.2
2·7
0·1
677
9.2
8·6
0·2
780
9.1
3·1
0·2
687
10.1
3·0
0·1
685
12.1
1·3
0·3
620
10.2
4·8
0·1
725
Central Diorite
Late Dacite Dike
3.1
8·1
0·3
775
2.1
2·7
0·1
675
3.2
8·2
0·3
776
3.1
1·3
0·1
620
5.2
6·3
0·2
750
4.1
3·2
0·1
690
5.1
7·3
0·3
765
4.2
7·2
0·4
763
7.1
2·8
0·1
680
5.1
3·5
0·1
697
8.1
7·9
0·5
772
7.1
3·3
0·1
693
11.1
6·2
0·3
749
8.1
3·7
0·1
703
11.2
6·0
0·4
746
10.1
2·1
0·1
656
12.1
7·2
0·3
763
12.1
2·4
0·1
668
14.1
8·4
0·1
779
17.1
1·8
0·1
645
17.1
7·8
0·3
772
JUNE 2012
et al., 1998). Despite some scatter, the entire dataset is remarkably uniform in that the d18OZrc values closely describe a unimodal normal distribution (Fig. 9). As such,
the weighted mean of 4·76 0·12ø (MSWD ¼ 2·0; 61 analyses) should be a good representation of the zircon population. Although an MSWD of 2·0 for the whole dataset of
d18OZrc might be indicating a scatter slightly in excess of
that expected from analytical error, the uniform distribution of the data does not reveal the existence of different
populations. Moreover, the absence of substantial differences between intrusive units and of any correlations with
the crystal sector analyzed or compositional and age parameters further precludes this possibility.
DISCUSSION
Magmatic vs hydrothermal origin of zircon
Teniente Porphyry
1.1
NUMBER 6
TZrc
estimates have been calculated with equations from
Ferry & Watson (2007) assuming an aSiO2 ¼ 1 and
aTiO2 ¼ 0·6.
single grain scale the d18O values are mostly within analytical uncertainty, with an intragrain variation lower than
0·8ø, whereas variations within samples of the different
units are up to 1·1^1·7ø. It is noteworthy that the El
Teniente d18OZrc average is within the lower limit of
normal mantle zircon values of 5·3 0·6ø (2s; Valley
A key question to be investigated in this study relates to the
nature of the zircon, whether it is of magmatic origin or if
there is a significant zircon fraction that can be interpreted
to be of hydrothermal origin. As such, zircon compositional and morphological features may be a consequence
of fluid circulation in hydrothermal systems with new
zircon growth and/or recrystallization. Diffusion of most
elements in zircon is unlikely, unless assisted by defects in
the crystal structure as in cracked or metamict zircons
(Cherniak & Watson, 2003). However, such areas can be
recognized as being altered through imaging techniques
and thus avoided during analysis.
The principal evidence for a hydrothermal imprint in
the El Teniente zircons comes from particular textural
and chemical features shown by several grains. Texturally,
they show a mottled external texture in transmitted light
images. Similar features have been recognized in other
hydrothermal deposits for zircons that chemically and isotopically preserve primary magmatic characteristics (van
Dongen et al., 2010). In El Teniente this surficial mottling
feature also corresponds to weakly luminescent high U^
Th rims (Fig. 3). The bimodal U^Pb age distribution
shown by zircon populations from the four older units
have been previously interpreted as representing the crystallization and alteration age of the respective intrusions
(Maksaev et al., 2004). This interpretation is also supported
by the agreement of the latter age with alteration, mineralization, and intrusive event ages obtained by different geochronological systems (Ar^Ar and Re^Os) on different
mineral phases. The younger units, the Teniente Porphyry
and the Late Dacite Dike, show unimodal zircon U^Pb
age distributions interpreted as representing the crystallization ages. Samples studied from all units correspond to
massive intrusive rocks that are heavily altered, as is characteristic for rocks within the deposit. Because the main
objective of this work was to study the genesis of magmas
related to porphyry copper mineralization, most of the
analyses were made on zircon spots inferred to be
1104
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Table 4: Lu and Hf isotopic data for zircons from El Teniente deposit
Spot
176
Hf/177Hf
2s
176
Lu/177Hf
2s
206
Pb/238U
2s
eHf(t)
2s
age (Ma)
2
TDM
(crustal)
A Porphyry
1.1
0·283001
0·000028
0·000256
0·000004
6·6
0·8
8·2
1·0
504
3.1
0·282965
0·000025
0·000300
0·000016
6·8
0·7
7·0
0·9
585
4.1
0·283000
0·000028
0·000415
0·000010
6·4
0·7
8·2
1·0
505
5.2
0·283010
0·000031
0·000801
0·000105
5·7
0·7
8·5
1·1
484
6.1
0·282988
0·000042
0·001838
0·000036
6·5
0·2
7·8
1·5
535
12.1
0·282988
0·000028
0·000337
0·000014
6·8
1·0
7·8
1·0
533
13.1*
0·282958
0·000034
0·000311
0·000008
6·4
0·2
6·7
1·2
601
13.2
0·282976
0·000026
0·000381
0·000016
6·2
0·7
7·3
0·9
561
1.1
0·282998
0·000048
0·001010
0·000058
6·4
0·3
8·1
1·7
512
3.1
0·282992
0·000028
0·000354
0·000025
5·8
0·6
7·9
1·0
525
3.2
0·283005
0·000028
0·000341
0·000016
6·3
0·7
8·4
1·0
494
4.2
0·282996
0·000025
0·000304
0·000011
6·3
0·8
8·1
0·9
515
7.1*
0·282998
0·000024
0·000236
0·000005
6·1
0·1
8·1
0·9
511
7.2
0·282996
0·000030
0·000615
0·000011
5·6
0·6
8·0
1·1
517
9.1*
0·282980
0·000058
0·000853
0·000040
6·0
0·3
7·5
2·0
553
9.2
0·282987
0·000024
0·000333
0·000008
6·3
0·5
7·8
0·9
535
10.1
0·282977
0·000026
0·000211
0·000002
6·2
0·3
7·4
0·9
559
Sewell Stock
Central Diorite
3.1
0·282974
0·000023
0·000300
0·000005
6·4
0·9
7·3
0·8
566
3.2
0·282966
0·000032
0·000210
0·000002
4·8
1·1
7·0
1·1
584
5.1
0·282978
0·000026
0·000314
0·000010
5·1
0·7
7·4
0·9
556
5.2
0·282984
0·000022
0·000168
0·000006
5·6
0·6
7·6
0·8
542
7.1
0·282995
0·000032
0·000299
0·000020
7·0
0·6
8·0
1·1
518
8.1
0·282947
0·000026
0·000319
0·000015
5·7
0·7
6·3
0·9
627
11.1
0·283001
0·000027
0·000323
0·000007
6·8
0·8
8·3
0·9
503
11.2
0·282995
0·000030
0·000465
0·000069
6·0
0·8
8·0
1·1
517
12.1
0·282980
0·000027
0·000726
0·000025
5·7
0·6
7·5
0·9
551
14.1
0·283001
0·000029
0·000393
0·000007
6·5
0·8
8·2
1·0
505
17.1
0·282995
0·000029
0·000560
0·000026
6·5
1·4
8·0
1·0
517
1.1
0·282980
0·000022
0·000369
0·000003
6·4
0·6
7·5
0·8
551
2.1
0·282986
0·000048
0·000699
0·000059
6·7
0·8
7·7
1·7
538
4.1
0·282982
0·000042
0·000483
0·000044
5·9
0·6
7·6
1·5
546
8.1
0·283000
0·000040
0·000480
0·000016
5·9
0·5
8·2
1·4
506
8.2
0·282991
0·000037
0·000621
0·000047
27·8
0·6
8·3
1·3
513
11.1
0·282985
0·000028
0·001204
0·000025
79·4
1·5
9·2
1·0
499
12.2
0·282980
0·000029
0·000927
0·000006
5·3
0·7
7·5
1·0
552
13.1
0·282987
0·000023
0·000238
0·000005
6·3
0·3
7·7
0·8
535
1.1
0·282947
0·000029
0·000504
0·000021
5·4
0·4
6·3
1·0
626
2.1
0·282975
0·000041
0·000577
0·000050
5·3
0·4
7·3
1·4
564
4.1
0·282968
0·000029
0·000459
0·000031
4·7
0·5
7·0
1·0
580
5.1
0·282973
0·000048
0·000625
0·000029
4·8
0·3
7·2
1·7
567
Northern Diorite
Teniente Porphyry
(continued)
1105
JOURNAL OF PETROLOGY
VOLUME 53
NUMBER 6
JUNE 2012
Table 4: Continued
Spot
176
Hf/177Hf
2s
176
Lu/177Hf
2s
206
Pb/238U
2s
eHf(t)
2s
age (Ma)
2
TDM
(crustal)
6.1
0·282960
0·000024
0·000479
0·000026
5·6
0·4
6·8
0·9
597
6.2
0·282985
0·000022
0·000430
0·000023
5·0
0·8
7·6
0·8
542
8.1
0·282952
0·000028
0·000678
0·000009
5·1
0·3
6·5
1·0
615
9.2
0·282978
0·000051
0·000340
0·000006
5·4
0·3
7·4
1·8
557
15.1
0·282982
0·000037
0·000537
0·000011
5·2
0·4
7·6
1·3
547
17.1
0·282996
0·000045
0·000592
0·000041
5·3
0·4
8·0
1·6
515
2.1
0·282956
0·000031
0·000577
0·000024
4·5
0·3
6·6
1·1
607
3.1
0·282945
0·000032
0·000760
0·000025
5·2
0·4
6·2
1·1
631
4.1
0·282951
0·000029
0·000449
0·000020
4·7
0·4
6·4
1·0
618
4.2
0·282968
0·000034
0·000280
0·000023
6·2
0·6
7·0
1·2
580
5.1
0·282947
0·000034
0·000876
0·000018
5·0
0·3
6·3
1·2
627
7.1
0·282957
0·000029
0·000729
0·000014
4·4
0·3
6·6
1·0
606
10.1
0·282968
0·000027
0·000388
0·000023
4·7
0·3
7·0
0·9
580
12.1
0·283002
0·000039
0·000398
0·000025
4·9
0·5
8·2
1·4
504
15.1
0·282962
0·000029
0·000497
0·000012
4·8
0·3
6·8
1·0
594
17.1
0·282995
0·000029
0·000560
0·000026
4·8
0·3
8·0
1·0
518
Late Dacite Dike
Söderlund et al. (2004) 176Lu decay constant of 1·867 1011 has been used in these calculations. For eHf(t) values the
chondritic values of Blichert-Toft & Albarède (1997) have been used along with the corresponding zircon spot age.
Two-stage depleted mantle model age TDM2 was calculated using the present-day depleted mantle values of Vervoort
& Blichert-Toft (1999) assuming a crustal average of 176Lu/177Hf ¼ 0·015 (Goodge & Vervoort, 2006).
magmatic in origin. However, it is still necessary to consider to what extent a hydrothermal imprint might have affected the chemical and isotopic characteristics of the
magmatic zircon grains, both for crystal sectors inferred
to be magmatic and for those that are suspected of being
affected by the hydrothermal imprint.
The CL images of the El Teniente zircons show preservation of a euhedral morphology and a well-developed
zoned structure, both common to simple zoned igneous
zircon (Fig. 3). Exceptions to this norm are the weakly luminescent, high U^Th rims developed around some
grains from the Sewell Stock and the A Porphyry (Fig. 3).
These rims are not associated with any other particular
compositional signature or U^Pb age group. Overall the
zircon populations show rather homogeneous morphological features that are distinctive for different intrusive
units. These characteristics argue against hydrothermal
recrystallization for most zircons and suggest that either
this process or new zircon growth is restricted to the limited development of overgrowth rims. Therefore, in general
the zircon grains from a single intrusion are compositionally homogeneous, and in terms of Y þ REE contents and
O^Hf isotopic composition they are indistinguishable between units. Discernible compositional variations arise in
the somewhat higher relative contents of U, Th and Hf
in zircons from the younger intrusive rocks, the Teniente
Porphyry and the Late Dacite Dike, and the general
trend of decreasing zircon Ti content with decreasing age
(Figs 6 and 7). Zircon-Ti thermometry records temperatures that are higher than the maximum estimated for
hydrothermal fluids during ore deposit formation (500^
6008C; Cannell, 2004). Overall, all these characteristics
are consistent with the zircons being formed by crystallization from a series of cogenetic melts at magmatic temperatures. Particular morphological and chemical features are
probably a first-order consequence of the particular melt
composition and variations in the magmatic evolution of
each intrusive unit.
The origin of the low d18OZrc in the El Teniente intrusive
rocks, whether magmatic or hydrothermal, has profound
implications in terms of petrogenetic processes. If this feature were produced by hydrothermal alteration of the
zircon grains, which are devoid of crystal defects, the process would have occurred through isotopic exchange by
volume diffusion inwards from grain boundaries. Watson
& Cherniak (1997) have shown experimentally that even
under wet conditions oxygen diffusion in zircon is extremely sluggish. For grains between 100 and 300 mm, the
range of El Teniente zircons, timescales to achieve complete isotopic exchange at 6008C are between 216 and
1106
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Table 5: d18O (VSMOW) of zircons from El Teniente
deposit
Spot
d18O
2s
Spot
(ø)
d18O
2s
(ø)
A Porphyry
Northern Diorite
1.1
5·13
0·58
1.1
5·37
0·60
3.1
4·59
0·62
2.1
4·89
0·59
4.1
5·13
0·63
4.1
5·63
0·60
5.2
4·34
0·62
6.1
4·61
0·59
6.1
4·47
0·57
8.1
5·43
0·59
7.1
5·52
0·59
8.2
4·62
0·61
11.1
4·85
0·57
11.1
4·88
0·61
12.1
5·51
0·59
11.2
4·38
0·62
13.1*
4·22
0·60
12.2
3·96
0·72
13.2
4·26
0·63
13.1
4·87
0·58
Sewell Stock
Fig. 8. Initial eHf isotope ratios in zircon grains relative to the
respective spot age showing the restricted range of values defined
by intrusive units from the El Teniente deposit. Chondritic (CHUR:
Blichert-Toft & Albare'de, 1997) and depleted mantle reservoirs
(Vervoort & Blichert-Toft, 1999) are shown for reference. Symbols as
in Fig. 5. Average analytical error bar at 2s is indicated.
Teniente Porphyry
1.1
3·64
0·60
1.1
4·52
0·74
1.2
5·25
0·61
4.1
5·53
0·77
3.1
4·57
0·61
5.1
5·39
0·75
3.2
4·83
0·65
6.1
4·83
0·74
4.2
4·72
0·61
6.2
5·57
0·76
7.1*
4·45
0·64
8.1
5·59
0·75
7.2
5·15
0·60
9.1
4·12
0·75
9.1*
4·76
0·56
9.2
4·68
0·76
9.2
5·01
0·57
15.1
4·73
0·73
10.1
4·47
0·60
17.1
4·96
0·73
Central Diorite
Late Dacite Dike
3.1
5·28
0·74
2.1
5·04
0·74
3.2
4·92
0·84
3.1
4·19
0·75
5.1
4·78
0·77
4.1
4·52
0·73
5.2
4·95
0·80
4.2
4·64
0·76
7.1
5·13
0·73
5.1
4·25
0·73
8.1
4·14
0·80
7.1
4·73
0·74
11.1
4·30
0·77
10.1
4·20
0·74
11.2
4·73
0·76
12.1
4·66
0·73
12.1
5·06
0·74
15.1
3·71
0·77
14.1
4·32
0·77
17.1
3·96
0·75
17.1
4·39
0·75
1945 Myr, respectively. Shorter timescales of 33^299 Myr
are obtained using the empirical determinations for wet
diffusion of Zheng & Fu (1998), but they are still unreasonably high when considering the timescales of hydrothermal
processes within the El Teniente deposit (Cuadra, 1986;
Maksaev et al., 2004; Cannell et al., 2005). Additionally,
partial exchange by diffusion is expected to produce zircons with isotopic zonation, and the process should proceed faster in smaller grains than in larger ones. Both
Fig. 9. Variation of d 18OZrc for the El Teniente deposit. (a) d18OZrc
relative to the respective spot age. Average analytical error bar at
2s is indicated. (b) Histogram of d 18OZrc values for all analyses
shown in (a) overlaid by a cumulative probability curve (calculated
with Isoplot version 3.00; Ludwig, 2003). Despite scatter, El Teniente
zircons show a simple unimodal distribution that extends to values
lower than those for normal mantle zircons (Valley et al., 1998).
1107
JOURNAL OF PETROLOGY
VOLUME 53
characteristics are absent from El Teniente. This strongly
suggests preservation of magmatic d18OZrc values regardless of hydrothermal alteration processes and, consequently, the data reflect a primary magmatic signature.
In summary, the morphological, chemical and isotopic
characteristics of the El Teniente zircons support the conclusion that the areas analyzed relate to the primary magmatic signature. A possible hydrothermal imprint is
evidenced by the presence of overgrowth rims and a
mottled texture, and in the bimodal U^Pb age distribution
in the older units. However, from the available data, the interpretation that the younger U^Pb ages represent hydrothermal alteration events relies heavily on their agreement
with ages obtained for gangue and ore minerals by other
geochronological methods (Ar^Ar and Re^Os; Maksaev
et al 2004).
Zircon records of magmatic evolution
As noted above, zircon morphological characteristics are a
first-order result of the melt composition from which they
crystallize, along with the temperature and cooling rate
during crystallization. With decreasing age the El Teniente
zircons show a progressive reduction of the {100} prismatic
form and an increase in the {110} form (Figs 3 and 4). In
the Teniente Porphyry the {100} form is poorly developed,
and it is almost completely absent in crystals from the
Late Dacite Dike (Figs 3 and 4). The size relations between
the two prismatic morphologies of zircon are strongly dependent on the chemical parameters of the growth
medium. Development of a {110}-dominated form is
favored by a growth blocking effect produced by the adsorption of protons, provided in the growth environment
by H2O molecules and hydrated complexes, and of elements such as U, Th, P and Y (Benisek & Finger, 1993;
Vavra, 1994), all of which are expected to increase in concentration with the degree of melt fractionation. In the El
Teniente zircons this mechanism controlling prism development is documented by the higher concentration of
elements as U, Th and Y shown by zircon populations
from the younger units (Fig. 6). This morphological and
chemical evolution reflects the progressively more differentiated nature of the younger intrusions. Additionally,
zircon crystals show a systematic increase in aspect ratio
with decreasing age (Fig. 4), a parameter that has been empirically shown to depend on cooling rate during crystallization. Thus, besides the decreasing crystallization
temperatures recorded by the Ti-in-zircon thermometry,
conditions of increasing cooling rate can be inferred from
the zircon aspect ratios. Evolution of these parameters proceeds in accordance with the progressive waning of igneous activity within the ore deposit. Additionally, strong
regional uplift and unroofing in the Andean range during
this period (Far|¤ as et al., 2008; Maksaev et al., 2009) are
probably also key factors controlling this evolution.
NUMBER 6
JUNE 2012
Despite the compositional differences observed between
the A Porphyry (SiO2 57%) and the remaining felsic
porphyries (SiO2 67%), all the zircons share similar
Hf^O isotopic compositions (Figs 8 and 9). This, along
with the general absence of older inherited zircon components, argues against significant crustal contamination in
the magmatic processes responsible for generating the different members of the suite. However, the El Teniente
range of initial eHf values between þ6·4 to þ8·4 records
an enrichment that, unless inherited from the source,
could result from crustal contamination processes during
the early stages of magma evolution. Recently published
data on the Hf isotopic composition of Cenozoic igneous
rocks from Central Chile show that these units share a
similar signature to that of El Teniente (Montecinos et al.,
2008; Deckart et al., 2010; Fig. 10). With the exception of intrusive units that are about 15 Ma in age, the Cenozoic
magmatism has remained remarkably uniform, with initial eHf values between þ6 and þ10 for just over 25 Myr.
This holds true despite the fact that it includes compositionally different rocks formed under contrasting tectonic
regimes and margin configurations (Fig. 10). The predominantly basic to intermediate igneous rocks from the
Oligocene^early Miocene Abanico Formation, formed in
an extensional setting over a thinned crust and with minimal crustal contamination, are indistinguishable from the
fractionated middle Miocene^Pliocene ones formed under
a contractional regime in a progressively thickening crust
(Fig. 10). This indicates that the control of Hf isotopic enrichment observed in El Teniente, and in the Cenozoic
magmatism in general, resides in the mantle source,
ruling out significant crustal contamination in their
genesis.
As discussed above, the coherent morphological, chemical and Hf isotopic characteristics of the El Teniente zircons indicates crystallization from a series of cogenetic
melts. These observations fully agree with the model of
Stern et al. (2010), who argued that the Late Miocene and
Pliocene plutonic rocks that host the deposit were derived
from a large, long-lived, thermally and chemically stratified, open-system magma chamber, or magmatic plumbing
system. Moreover, this hypothesis has also been favored in
explaining the structural patterns and the chemical evolution of the hydrothermal systems within the deposit
(Cannell et al., 2005; Klemm et al., 2007). The El Teniente
d18OZrc weighted mean of 4·76 0·12ø is considered to
be a primary igneous feature, and therefore might result
from crystallization from low-18O magmas. In the context
of El Teniente, sourced by a long-lived magmatic system
with an extensive record of hydrothermal activity, a likely
scenario to produce this composition is the assimilation of
hydrothermally altered wall-rocks. Simple mixing calculations indicate that the average of d18OZrc 4·7ø could be
produced from a mantle-derived magma by 8^11% bulk
1108
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Fig. 10. (a) Zircon initial eHf isotope ratios of Oligocene^Pliocene igneous rocks from Central Chile. Gray shaded field highlights the relatively
homogeneous Hf-isotopic signature shown by Central Chilean magmatism during the last 25 Myr. Legend: 1, Abanico and Farellones formations NE of the R|¤o Blanco^Los Bronces deposit (Montecinos et al., 2008); 2, San Francisco batholith; 3, Yerba Loca pluton; 4, Cerro Meso¤n
Alto stock; 5, La Gloria pluton (Deckart et al., 2010); 6, El Teniente intrusive rocks; 7, El Teniente inherited cores. (b) Schematic representation
of the main tectonic events in the Andean range for the studied region during the time span considered in (a). Vertical axis shows the Andean
range segmented, from west to east, according to the main morphostructural units that form it. The main tectonic events illustrated include extensional tectonics related to Abanico Basin development, compressional tectonics related to basin inversion and subsequent eastward migrating
shortening episodes, and the main Andean uplift event. Data from Giambiagi & Ramos (2002) and Far|¤ as et al. (2010). All locations and the distribution of Andean morphostructural units are indicated in Fig. 1.
assimilation of crustal material with d18OWR of between ^2
and 0ø. Although limited, such an amount of crustal contamination restricts possible contaminants to the nearly
6 km thick sequence of Oligocene^Miocene igneous rocks
into which the deposit is emplaced. If contaminated by
Mesozoic or Paleozoic basement rocks, such as those that
crop out near the Chilean^Argentinean border and that
probably underlie the whole region (Fig. 2), then the El
Teniente Hf isotopic composition would be shifted towards
radiogenic values inconsistent with those observed.
Wall-rock assimilation under upper crustal conditions can
proceed either by assimilation^fractional crystallization
(AFC)-type processes (DePaolo, 1981) or by the incorporation of a low-degree partial melt (Campbell & Turner,
1987; Huppert & Sparks, 1988). However, in El Teniente,
characteristics such as the lack of zircon intragrain d18O
variation and the homogeneous d18OZrc values shown by
the intrusive rocks irrespective of the age of intrusion or
composition strongly argue against AFC-type processes.
A second possibility involves the formation of a magma
layer by melting of the roof rocks of an underlying basaltic
magma chamber that provides the heat for melting
(Campbell & Turner, 1987; Huppert & Sparks, 1988).
Models show that the melted material will remain at the
1109
JOURNAL OF PETROLOGY
VOLUME 53
top of the chamber and be chemically isolated from the
basaltic magma at the bottom, thus separating assimilation from crystallization in space and time. This process
predicts the formation of granodiorite or tonalite melts
and can occur at upper levels of the crust only where the
rocks have been pre-heated by earlier intrusions.
Although this scenario seems likely in the case of the El
Teniente magmatism, this process is not consistent with
two main observations: (1) the general absence of inherited
zircons or zircon cores; (2) the simultaneous generation of
melts of dacitic to andesitic composition which also have
the same O isotopic signature. Although the occurrence of
this process cannot be ruled out, it seems unlikely to be
fully responsible for generating the diverse magmas with
low-O isotopic compositions at El Teniente.
Another process capable of lowering the magma d18O
signature is through fractionation of O isotopes during
degassing. However, high-temperature magma volatile
loss is expected to produce a relatively minor shift towards
lower d18O values in the resulting melt in the case of
a pure H2O volatile phase, and even lower shifts are
expected in the case of SO2 (Eiler, 2001). For example,
high-temperature (46008C) melt-volatile fractionation
predicts that 10% degassing of pure H2O from a gabbroic
melt would produce a degassed melt with a d18O value of
0·11ø lower than the original one, and no fractionation is
predicted for more felsic compositions such as granodioritic or granitic melts (Zhao & Zheng, 2003). Thus, this process seems unable to reproduce the values observed at El
Teniente, which are on average 0·6ø lower than would be
expected for zircon from a mantle-derived magma. Thus,
as for Hf, the El Teniente O isotopic composition is
inferred to be a characteristic inherited from the mantle
source.
Magma generation: constraints on the
source and melting processes
A major reconfiguration of the Chilean continental
margin took place during the Cenozoic. The Oligocene^
early Miocene extensional Abanico Basin was inverted at
around 21 Ma, and this was followed by a contractional
regime in which shortening, thickening and uplift characterized the constructive period of the Andean orogen
(Charrier et al., 2002, 2009; Giambiagi & Ramos, 2002;
Kay et al., 2005; Far|¤ as et al., 2008, 2010). Whether directly
related or not, Nazca and South American plate convergence parameters also varied during this period.
Obliqueness abruptly diminished from 45^558 to 15^
208 around 26 Ma, and the convergence rate increased
from 6 cm a1 at 27 Ma to reach 12^20 cm a1 around
15 Ma and then decreased steadily to its present value
of 8 cm a1 (Pardo-Casas & Molnar, 1987). The limited
variation in Hf isotopic composition shown by Central
Chilean igneous rocks throughout this period strongly suggests buffering by a stable isotopic reservoir. Hf depleted
NUMBER 6
JUNE 2012
mantle model ages of 500^600 Ma highlight the enriched
nature of this reservoir. The subcontinental lithospheric
mantle and the lower crust constitute reservoirs able to
constantly imprint such a signature in magmas irrespective of the margin configuration and thus are likely to be
responsible for the observed isotopic composition of
Chilean Cenozoic igneous rocks. The Hf isotopic data for
El Teniente and other intrusive rocks of the region
(Deckart et al., 2010) agree with and expand the results of
Montecinos et al. (2008). Based on Pb, Sr and Hf isotope
data, those workers showed that there are no significant
variations in the Oligocene^middle Miocene magmatism
and attributed the enriched isotopic signatures to a characteristic inherited from the subcontinental lithospheric
mantle. Overall the Hf isotopic composition of the
Cenozoic magmatism in Central Chile can be considered
to have been derived from extensive and long-lived
MASH-type processes. Such MASH processes were originally proposed by Hildreth & Moorbath (1988) to explain
the chemical and isotopic variations in volcanic rocks
along the Chilean Southern Volcanic Zone. MASH domains are deep lithospheric zones where ascending, subduction-related, mantle-derived magmas are hybridized
generating relatively homogeneous magmas with chemical
and isotopic characteristics specific of the MASH zone
from which they evolved. An exception to the observed
homogeneous Hf isotopic compositions in Central Chile is
the higher initial Hf signatures (more depleted) recorded
by intrusive units at around 15 Ma (Fig. 10). These correspond to the Yerba Loca stock and the oldest portions of
the San Francisco Batholith (locations indicated in Fig. 2;
Deckart et al., 2010). This period coincides with an important change in the structural evolution of the Andean
orogen. Following Abanico Basin inversion, contractional
deformation that was mostly concentrated in the western
slope of the Andes migrated to the east, forming the
Aconcagua Fold and Thrust Belt in the Eastern Principal
Cordillera (Fig. 2; Giambiagi & Ramos, 2002). We suggest
that during this reorganization a transient period of stress
in the Andean range permitted rapid ascent of subduction-related, mantle-derived magmas that had little interaction with the upper lithosphere. Such a process is
supported not only by the high initial eHf values recorded
in the Yerba Loca stock, but also in the change towards
less depleted compositions recorded in the San Francisco
Batholith, with eHf values of þ8·5 to þ10·9 at 14 Ma and
þ7·0 to þ8·6 at 11Ma (Fig. 10).
As an alternative to the MASH model of Hildreth &
Moorbath (1988), Stern (1991) proposed source contamination to explain the variable enrichment of the Chilean
Southern Volcanic Zone magmas. In this model, incorporation into the mantle wedge of different amounts of subducted sediment and Paleozoic upper crust derived from
forearc subduction erosion could be responsible for the
1110
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
observed isotopic signatures of the arc magmas. This process was also proposed to in part explain the adakite-like
chemistry of the El Teniente intrusive rocks within the context of the ‘normal’ Cenozoic magmatism in Central Chile
(Kay et al., 2005). According to this model, crustal blocks
incorporated into the mantle wedge at peak times of subduction erosion during continental margin evolution are
subjected to high-pressure metamorphism and partial
melting, and generate the adakite-like magmas later
emplaced into the upper plate. However, if such a process
were responsible for the chemical differences between the
El Teniente magmas and the preceding magmatism it
would undoubtedly result in a distinct isotopic signature
between them, which is not the case as recorded by the
Hf isotopic data (Fig. 10). Several lines of evidence, other
than just geochemical, support subduction erosion as a
long-term process occurring along the Chilean continental
margin (Stern, 1991; Laursen et al., 2002; Kay et al., 2005).
However, the extent to which this process controls the
chemical and isotopic signatures of Andean arc magmas
is certainly variable. Subduction erosion products may
enter the asthenospheric source of the Cenozoic magmas,
but their impact on magma geochemistry is probably overwhelmed by the later imprint that they are subjected to in
deep lithospheric MASH zones.
In recent years, several workers have highlighted the
possibility of melting subduction-modified lithosphere as a
process for generating porphyry Cu Mo Au related
magmas (Kay & Mpodozis, 2001; Richards, 2009; Shafiei
et al., 2009). Arc magmatism is ultimately a means of material transfer from the oceanic slab and the asthenospheric mantle to the overriding plate. The interaction
with, and differentiation of, these magmas deep in the
lithosphere results in the formation of hydrous cumulate
zones, of mafic to ultramafic composition, where significant accumulation of amphibole and the presence of residual sulfide phases can account for water and metal
storage in the upper lithosphere (e.g. Davidson et al., 2007;
Jagoutz et al., 2007; Richards, 2009). Thus, after long periods of arc magmatism, the lithospheric roots are expected
to be a fertile reservoir from which potentially ore-forming
magmas may be extracted.
During the Cenozoic evolution of the Central Chilean
Andes, progressive lithospheric thickening has been
inferred to involve a change in mineralogy in the deep
lithosphere from low-pressure amphibole-dominated to
higher pressure garnet-dominated assemblages. This is evidenced in the chemical trends defined by the arc-related
igneous rocks formed during this period (Kay &
Mpodozis, 2001, 2002; Kay et al., 2005). Such a process involves the generation of an H2O-undersaturated melt, the
H2O being provided by the dissociation of hydrous mineral phases. This dehydration melting reaction overall involves Hornblende þ Plagioclase Quartz ! Hornblende
þ Pyroxene þ Garnet þ Melt, and is basically the metamorphic transition from amphibolite to eclogite facies
(Sen & Dunn, 1994; Wolf & Wyllie, 1994). Experimental
work has shown the ability of this process to generate hydrous silicate melts of felsic to intermediate composition,
which share many chemical similarities with igneous
rocks of the tonalite^trondhjemite series and adakites.
Adakite-like characteristics such as intermediate to high
SiO2 content, highly fractionated REE patterns, trondhjemitic affinities, and high concentrations of Sr and high Sr/
Y ratios, have all been recognized as characteristic of the
El Teniente intrusive rocks (Rabbia et al., 2000; Kay et al.,
2005; Stern & Skewes, 2005; Stern et al., 2010), as well as of
many other ore deposit-related igneous rocks worldwide.
Moreover, the Os isotope compositions of sulfides in
Chilean porphyry copper deposits, including El Teniente,
suggest significant crustal residence of the Os and thus a
probable addition from the lower crust (Mathur et al.,
2000).
A key question is the ability of the crust at the time
of formation of the El Teniente magmas to stabilize garnet through dehydration melting of basic precursors.
Experimental studies have reproduced this reaction over a
wide range of pressures and temperatures from 10 to
16 kbar and from 700 to 10008C (Wolf & Wyllie, 1993,
1994; Wyllie & Wolf, 1993; Sen & Dunn, 1994; Rapp &
Watson, 1995; Lo¤pez & Castro, 2001), in agreement with
observations from natural examples (e.g. Garrido et al.,
2006; Berger et al., 2009). In terms of pressure, this indicates
a minimum crustal thickness of 30^35 km. However, regarding temperature there is a thermal barrier to overcome to drive this process, as, assuming a conservative
thermal gradient of 208C km1, the base of such a crust
would reach no more than 600^7008C. In Central Chile,
the increasing La/Yb ratios of the Cenozoic igneous
rocks, particularly in the El Teniente region, suggest the
increasing involvement of garnet as a high-pressure residual assemblage (Kay et al., 2005). Moreover, the current
45^50 km crustal thickness under this area must have
been reached no later than 4 Ma, the time at which uplift
of the Andean orogen to its current altitude was completed
(Far|¤ as et al., 2008, 2010). Uplift has been inferred to be primarily the result of an isostatic crustal response to tectonic
shortening and thickening processes. Thus, the chemical
trends and tectonic evolution of the margin during the
Cenozoic both indicate the presence of garnet in lower
crustal assemblages. Less straightforward is how to elucidate the coeval thermal regime governing this zone,
which would ultimately condition the occurrence of dehydration melting reactions.
Since the original MASH model proposed by Hildreth
& Moorbath (1988), several studies have focused on modeling the dynamics of magmatism-driven processes occurring at or close to the Moho in convergent margin settings
1111
JOURNAL OF PETROLOGY
VOLUME 53
(Petford & Gallagher, 2001; Annen & Sparks, 2002; Annen
et al., 2006). Annen et al. (2006) developed a comprehensive
model built upon the concepts of underplating and
high-pressure basalt differentiation, incorporating aspects
of AFC- and MASH-type processes, to evaluate the effects
of repeated basalt intrusion into the crust. Their work proposed the existence of ‘deep crustal hot zones’, which
result from repeated deep intrusion of mantle-derived, hydrous, basalt sills, either by injection at a fixed depth at
the Moho or randomly throughout the lower crust. This
leads to a scenario in which evolved melts reaching
middle to upper crustal levels are generated from
H2O-rich parental basalts, both by partial crystallization
of the basalts themselves and by partial melting of the surrounding crustal rocks through heat and H2O transfer
from the cooling basalts. Further mixing between these
two end-members can create a large range of intermediate
and silicic melts with variable composition. Moreover,
models indicate that ‘deep crustal hot zones’ are the place
where much of the geochemical diversity of magmas originates, owing to substantial variations in melt proportions
and temperature at such depths. Either at a fixed or
random depth, repeated basalt intrusion induces a significant thermal perturbation into the lower crust and below
the Moho, as a result of heat transfer into the country
rocks, which will take several millions of years to decay
(Annen & Sparks, 2002). Thermal gradients in both situations would allow the lower sections of a 30 km thick
crust to reach temperatures much higher than 7508C
(Fig. 11). These are high enough to drive dehydration melting of amphibolite in the stability field of garnet at pressures 410 kbar (Fig. 11). Additionally, the solidus curve of
this reaction has an abrupt backbend towards lower temperatures that coincides with the garnet-in curve towards
higher pressures (Fig. 11). According to different models,
the melting temperature is lowered by 100^2008C once
entering the stability field of garnet (Wyllie & Wolf, 1993;
Wolf & Wyllie, 1994; Rapp & Watson, 1995; Lo¤pez &
Castro, 2001). An amphibolitic lower crust can reach temperatures up to 750^8708C at pressures 510 kbar without
melting, but undergo dehydration melting in response to
an increase in crustal thickness (Fig. 11). These observations
can certainly vary in terms of detail when considering the
uncertainties related to the Annen et al. (2006) model, as
well as those related to the phase relationships. However,
the overall scenario is unlikely to change: repeated basalt
intrusion at or close to the Moho will induce a thermal perturbation that will significantly widen the field in which
dehydration melting can occur at the base of a 433 km
thick crust (Fig. 11). Regarding El Teniente, continuous
and extensive magmatic activity in Central Chile preceded
the formation of the ore deposit for at least 30 Myr; thus
a perturbed thermal gradient may have been established
in the crust well before ore deposit formation. Moreover,
NUMBER 6
JUNE 2012
Fig. 11. Pressure^temperature diagram showing the solidus for dehydration melting of amphibolite and the volume per cent of melt
involved in this reaction (gray dashed lines; Lo¤pez & Castro, 2001).
A characteristic of the amphibolite solidus is the abrupt back bend towards lower temperatures at pressures greater than 10 kbar (Wyllie
& Wolf, 1993; Wolf & Wyllie, 1994; Rapp & Watson, 1995; Lo¤pez &
Castro, 2001). Different geothermal gradients are indicated with
dashed and dotted black lines. Linear gradients of 20 and 308C km1
were calculated assuming a geobarometric gradient of 0·33 bar km1;
perturbed gradients from long-term basalt injection, at fixed and
random depth, at the base of an initially 30 km thick crust are from
Annen et al. (2006). Grt, garnet.
if high enough temperatures (750^8708C) were reached in
the lower crust before ore deposit formation, it could have
undergone dehydration melting following an increase in
crustal thickness and once the residual assemblages began
to stabilize garnet.
Two main processes can be invoked to explain the primary low-d18O composition of the El Teniente magmas:
(1) differentiation from or involvement of low-d18O material in the source; (2) isotopic fractionation during magma
generation. Regarding the former, low-d18O material in
the MASH source would be expected to similarly influence the isotopic composition of the Cenozoic magmatism
in the region. Currently there are no oxygen isotope data
for these rocks; however, it is unlikely that they all share a
low-d18O signature. The second possibility implies partial
melting in a deep lithospheric MASH zone in which all
Cenozoic magmatism has evolved. During partial melting
fractionation of oxygen isotopes can be considerable and
would have measurable effects on the generated melts
(Eiler, 2001). To evaluate this possibility within a dehydration melting process a simple model has been constructed
following the calculation approach of Eiler (2001) using
the experimental results of Wolf & Wyllie (1994) and
Getsinger et al. (2009). Wolf & Wyllie (1994) studied the
progressive change in mineral assemblage and mineral
and melt chemistry during dehydration melting of a natural low-K amphibolite (67·4% hornblende þ 32·5%
plagioclase; wt % SiO2 ¼ 48·4%), at 10 kbar between 850
and 10008C. Getsinger et al. (2009) studied the impact on
melt composition and residual mineral assemblage produced by melt segregation and accumulation during
1112
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
dehydration melting of an initial composition of a natural
slightly alkaline metabasalt (45% hornblende þ 25%
plagioclase þ10% biotite þ10% epidote þ 2% clinopyroxene, plus minor rutile, titanite and magnetite; wt %
SiO2 ¼ 46·4%), at 14 kbar between 925 and 10008C. An
initial d18OWR ¼ 5·8ø was assumed for the system,
which corresponds to that of a mantle-derived basaltic
magma (5·5ø) contaminated with 3% subducted sediment (15ø; Eiler, 2001), considering that a deep MASH
zone would ultimately have the O isotope signature of
subduction-related mantle-derived magmas. Results are
displayed in Fig 12, where the calculated d18O for the
whole system initial composition, residue, generated melt,
and zircon in equilibrium with the latter are plotted
against melt wt % SiO2 (further details of this calculation
are given in the figure caption and the Appendix). Results
from the two experiments are not directly comparable,
but are complementary. The Wolf & Wyllie (1994) experiments reproduce an incremental reaction; incremental
heating of a fixed initial composition overall produces
increasing amounts of melt with a progressively more
basic composition in each temperature step (Fig. 12). For
the Getsinger et al. (2009) experiments the initial composition varies; different proportions of the original metabasalt are mixed with a 15% partial melt of this same
composition forming composites that are then melted at
different temperatures (Fig. 12). Models for both experiments show that dehydration melting processes can generate melts, and thus zircons that would crystallize from
them, with low d18O signatures and with compositions
and d18OZrc in the range of those from El Teniente
(Fig. 12). How O-depleted the calculated zircons are relative to the d18O of a normal mantle zircon depends largely
on the compositional and temperature parameters. The
Fig. 12. Calculation of d18O for the components involved in experimental dehydration melting of a mafic amphibolite (Wolf & Wyllie, 1994;
Getsinger et al., 2009) with a bulk oxygen isotope composition of 5·8ø. This value was assumed to be representative of a mantle-derived basaltic
magma (5·5ø) contaminated with 3% subducted sediment (15ø). The d18O of the whole system initial composition, residue, generated melt,
and zircon in equilibrium with the latter for each experimental step are plotted against melt wt % SiO2. Calculations for the melt and residue
were made following Eiler (2001) assuming a d18O for the whole system and considering the melt as the weighted sum of its normative mineralogy. Oxygen isotope fractionation was calculated using the empirical factors of Zheng (1991, 1993a, 1993b) and the corresponding temperatures
for each experimental step. d18OZrc for zircon in equilibrium with each generated melt was calculated according to the relation of this value
with the whole-rock d18O and wt % SiO2 determined by Valley et al. (1994). Further details on the calculations are given in the Appendix.
1113
JOURNAL OF PETROLOGY
VOLUME 53
lower the melt^residue compositional difference the lower
the 18OMelt^Residue and the generated melts will have a
d18O signature more similar to that of the residue and the
whole system; in other words, progressively more basic
melts will have less exotic low-d18O compositions (Fig. 12).
Additionally, such melts are overall produced at increasingly higher temperatures. The comparatively lower d18O
melts and zircons obtained from the Getsinger et al. (2009)
experiments, with respect to those of Wolf & Wyllie
(1994), also result from differences in these parameters.
Melts similar in composition from both studies (wt %
SiO2 67%, Fig. 12) are produced at higher temperatures
in the Getsinger et al. (2009) experiments. This causes a significant reduction of the 18OMelt^Residue and thus the
melts are comparatively more depleted in their oxygen isotope signature (Fig. 12), an effect that propagates to the
higher silica, lower temperature melts. This cannot be
associated directly with the effects of melt accumulation
and segregation, as the Getsinger et al. (2009) experiments
produce siliceous, high-temperature melts following the
initial 15% melting of the original alkali basalt. In summary, intermediate to felsic melts generated through dehydration melting processes will most probably have a
low-d18O signature, although the extent of depletion is
strongly dependent on temperature and compositional parameters during melting. Among equivalent compositions,
this will be more pronounced for melts generated at
higher temperatures. A comparatively less depleted signature is expected in melts of similar composition but
formed through fractional crystallization of more basic
precursors derived from dehydration melting processes.
Overall, in both experiments and in all melt compositions,
the effect on the d18O of the residue is minimal. Thus,
unless melt extraction is by a Rayleigh distillation process,
no measurable effect is expected on subsequent ascending
magmas that may interact with this residue.
M A G M AT I C M O D E L A N D F I N A L
REMARKS
The model described above aims to evaluate the O isotopic
composition of melts generated by dehydration melting reactions. It is not considered to be a close approximation to
magma genesis at El Teniente as it certainly does not take
into account the numerous complexities that can be
induced by other processes occurring in a MASH source.
Following the original model of Hildreth & Moorbath
(1988), we envisage that Cenozoic magmatism in Central
Chile originates from a deep lithospheric MASH reservoir,
in which ascending subduction-related, mantle-derived
magmas initially stall, isotopically homogenize, and differentiate until the resultant H2O-rich residual melts with a
lower density are able to continue their ascent to upper
crustal levels (Fig. 13). This model relies entirely on the
NUMBER 6
JUNE 2012
nearly constant Hf isotopic composition shown by the
Cenozoic igneous units of the region, for which calculated
depleted mantle model ages of 480^630 Ma indicate a
significant crustal residence time. Additionally, the
observed Hf isotope signature remains the same within
rocks of different composition that were formed at different
times, during a period when convergence parameters
varied and major changes in upper plate tectonics took
place. Residual assemblages, along with the original material in the MASH reservoir, can stabilize garnet through
a dehydration melting reaction to produce the fertile
magmas later involved in the El Teniente mineralization
(Fig. 13). This arises as a consequence of the thermal perturbation induced by the nearly 30 Myr of magmatism
that precedes the El Teniente ore deposit formation and
the increase in crustal thickness. Although this process produces an H2O-undersaturated melt, greater quantities of
H2O can be involved if the heat source efficiently fluxes
the source region with H2O (Annen et al., 2006).
Additionally, higher H2O contents can also result from
mixing with H2O-rich residual melts in the source or in
magma reservoirs at upper crustal levels, a most likely
scenario for the El Teniente magmas, which are inferred
to be related to long-lived magmatic chamber processes
(Stern et al., 2010). The current active volcanic zone in
Central Chile is located nearly 35 km east of El Teniente,
which along with the record of progressively younger igneous units towards the east reflects Cenozoic arc migration
along the margin (Stern & Skewes, 1995; Kay et al., 2005).
As early as 8 Ma magmatism had already reached the
area near the current active volcanic zone (Mun‹oz et al.,
2009), whereas volcanic activity was progressively waning
in the El Teniente region. Although there is no precise estimation of when arc magmas no longer reached the area
beneath El Teniente, by the time of ore deposit formation
this activity was probably in decline (Fig. 13). This might
have also influenced the fertility of the magmas leaving
the MASH source, by increasing the melt component
derived from dehydration melting reactions in this
domain and decreasing the component derived from primary basalt differentiation.
Even though El Teniente is one of the largest porphyry
copper deposits known at present, its formation is not
unique within the evolution of the Chilean continental
margin, which contains several other world-class deposits.
These are ultimately linked to a long history of
subduction-related magmatism, which, in the appropriate
conditions, leads to their formation. However, the specific
characteristics that make some areas especially productive
in forming numerous world-class deposits in a particular
metallogenic epoch is a question yet to be resolved.
This is the case for the northern Chile late Eocene^early
Oligocene and the Central Chile late Miocene^early
Pliocene Cu^Mo belts. Although emplaced at different
1114
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
Fig. 13. Schematic model for the generation of the El Teniente ore deposit fertile magmas within the context of the Cenozoic evolution of the
Central Chilean Andes. Magmatism from 28 Ma to 6 Ma occurring in the Western Principal Cordillera proceeded with the establishment
of a long-lived lower crustal MASH zone, which became progressively enriched through early differentiation of ascending subduction-related,
mantle-derived magmas. Following basin inversion, after 21Ma, coeval progressive crustal shortening and thickening induced dehydration
melting of the fertile cumulate residual assemblages of the MASH zone. This occurred upon reaching a critical crustal thickness and was
prompted by the perturbed thermal gradient induced by the nearly 30 Myr of preceding magmatism. Coevally, owing to arc migration, the
volume of primary magmas reaching the MASH zone beneath El Teniente probably decreased. This may also have influenced the fertility of
the magmas, by increasing the melt component derived from dehydration melting and decreasing the component derived from primary basalt
differentiation. These magmas, after subsequent ascent and differentiation at upper crustal levels, led to the El Teniente porphyry copper deposit
formation. Magmatic activity in the Western Principal Cordillera gradually ceased with progressive eastward arc migration, which is finally
fully established in the current location of the volcanic arc along the Eastern Principal Cordillera. Uplift and denudation processes during the
construction of the modern Andean orogen ultimately exhumed the deposits to the actual surface exposure. WPC, Western Principal
Cordillera; EPC, Eastern Principal Cordillera; FC, Frontal Cordillera; Fr, Foreland; CC, continental crust; OC, oceanic crust; SCLM,
sub-continental lithospheric mantle.
1115
JOURNAL OF PETROLOGY
VOLUME 53
times and through lithologically different sections of the
crust (Mpodozis & Ramos, 1989; Camus, 2003; Ramos
et al., 2004), they are formed at the end of periods of similar
geological evolution; that is, the time after extended
arc-related magmatism, during overall waning of igneous
activity and cessation of volcanism, and following stages
of strong compressive deformation, crustal thickening,
uplift and denudation (e.g. Maksaev & Zentilli, 1988;
Richards et al., 2001; Tosdal & Richards, 2001). Overall,
porphyry-related magmas share a similar composition.
The recurrence of deposit formation in different geological
frames but following similar histories indicates that they
result from the operation of regional processes rather than
singularities. In the context of progressive crustal thickening, dehydration melting of a long-lived MASH reservoir
can produce highly fertile magmas and/or enhance the fertility of subduction-related, mantle-derived magmas,
which can be further enhanced during middle^upper crustal magma chamber processes. Generation of fertile
magmas would reduce the high volumes necessary to
source metals into at least the giant deposits (Cline &
Bodnar, 1991; Richards, 2005; Stern & Skewes, 2005; Stern
et al., 2010). However, this constitutes one favorable, but
not a key aspect, of a multi-variable process, as several
other appropriate conditions also need to be met for their
final formation (e.g. Burnham, 1979; Carroll &
Rutherford, 1985; Tosdal & Richards, 2001; Cloos, 2002;
Richards, 2003, 2005). In this regard, for example, during
the formation of the northern Chile Paleocene^Early
Eocene metallogenic belt the crust never reached more
than 40 km in thickness (for a review see Camus, 2003)
and yet several comparatively minor porphyry copper deposits were formed. Other aspects of the geological evolution during this time differentiate this belt from the more
productive late Eocene^early Oligocene and late
Miocene^early Pliocene belts. However, they may represent examples of how, given the appropriate conditions, dehydration melting processes can enhance magma fertility
in different metallogenic epochs.
(3)
(4)
(5)
(6)
NUMBER 6
JUNE 2012
main igneous zircon populations in all other compositional parameters.
El Teniente zircons qualitatively describe an evolutionary trend towards more differentiated compositions, in terms of higher incompatible element
enrichment and increasing cooling rates in the progressively younger intrusions. This trend is accompanied by decreasing crystallization temperatures, as
measured by Ti-in-zircon thermometry. Such characteristics in the evolving magmatic system agree well
with the evolution of the region during this period as
characterized by increased uplift and denudation processes during the constructive period of the Andes
(Far|¤ as et al., 2008; Maksaev et al., 2009).
Zircon Hf and O isotopic composition are uniform at
the grain and sample scale and define a single signature for all El Teniente intrusions. This observation indicates a primary control from the source as opposed
to any significant crustal contamination processes
involved in magma genesis. Hf isotopic composition
is inferred to fingerprint the source and O the isotope
composition of the melting process.
Cenozoic magmatism in Central Chile, including that
at El Teniente, shows a remarkably homogeneous
Hf isotopic composition over a period of more than
25 Myr. Throughout this time the continental margin
was subjected to different configurations and tectonic
regimes, indicating that a stable and long-lived
MASH-type reservoir in the deep lithosphere has buffered the observed compositions.
Dehydration melting reactions in the fertile MASH
source probably occur in response to crustal thickening prompted by the anomalous thermal regime
governing this zone as a consequence of the long-lived
preceding magmatism. This process, coupled with
waning arc activity, is likely to generate fertile
magmas and/or enhance the fertility of subductionrelated mantle-derived magmas whose later evolution
in the upper crust can lead to the formation of porphyry copper deposits on a regional scale.
CONC LUSIONS
(1) The study of zircon populations from porphyryrelated igneous rocks provides valuable information
on petrogenetic processes that are otherwise obscured
by the intensive hydrothermal alteration that characteristically affects these rocks.
(2) Zircons from mineralization-related intrusive rocks
from the El Teniente porphyry Cu^Mo deposit show
morphological and compositional features that indicate crystallization from a series of cogenetic melts.
Some grains show evidence of a minor hydrothermal
imprint in high U^Th weakly luminescent overgrowth rims, which are indistinguishable from the
AC K N O W L E D G E M E N T S
We thank the geologists Patricio Zu¤n‹iga, Ricardo Floody
and Jose¤ Seguel from the Superintendencia de Geolog|¤ a El
Teniente, CODELCO-Chile, for providing mine access
and logistical support for the development of this work.
Help in organizing visits to El Teniente and assistance
during the field work of Marcela Cereceda and Rene¤
Padilla is gratefully acknowledged. We thank Dr G. Yaxley
and the technical staff of the Research School of Earth
Sciences, Australian National University (RSES-ANU),
and technical staff of the Departamento de Geolog|¤ a,
Universidad de Chile, for their helpful assistance in
1116
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
sample preparation and during the analytical sessions.
We also thank the reviewers J. P. Richards and
M. Chiaradia, and the Editor G. Wo«rner, whose comments
and discussions have helped us greatly to improve the
final paper.
FUNDING
This work was supported by the Chilean government
through the Comisio¤n Nacional de Ciencia y Tecnolog|¤ a^
CONICYT (Anillo ACT-18 project, PBCT program).
Additional funding was provided by the Departamento de
Postgrado y Post|¤ tulo, Universidad de Chile. This work is
part of the PhD thesis of M. Mun‹oz, which was supported
by a 4 years grant from CONICYT.
R EF ER ENC ES
Annen, C. & Sparks, R. S. J. (2002). Effects of repetitive emplacement
of basaltic intrusions on thermal evolution and melt generation in
the crust. Earth and Planetary Science Letters 202, 937^955.
Annen, C., Blundy, J. D. & Sparks, R. S. J. (2006). The genesis of intermediate and silicic magmas in deep crustal hot zones. Journal of
Petrology 47, 505^539.
Antofagasta plc. (2009). Annual Report and Financial Statements. London:
Antofagasta plc, 120 p.
Ballard, J. R., Palin, J. M. & Campbell, I. H. (2002). Relative oxidation states of magmas inferred from Ce(IV)/Ce(III) in zircon: applications to porphyry copper deposits of northern Chile.
Contributions to Mineralogy and Petrology 144, 347^364.
Benisek, A. & Finger, F. (1993). Factors controlling the development of
prism faces in granite zircons: a microprobe study. Contributions to
Mineralogy and Petrology 114, 441^151.
Berger, J., Renaud, C., Lie¤geois, J.-P., Mercier, J.-C. C. &
Demaiffe, D. (2009). Dehydration, melting and related garnet
growth in deep root of the Amalaoulaou Neoproterozoic magmatic
arc (Gourma, NE Mali). Geological Magazine 146, 173^186.
Blichert-Toft, J. & Albare'de, F. (1997). The Lu^Hf isotope geochemistry of chondrites and the evolution of the mantle^crust system.
Earth and Planetary Science Letters 148, 243^258.
Burnham, C. W. (1979). Magmas and hydrothermal fluids.
In: Barnes, H. L. (ed.) Geochemistry of Hydrothermal Ore Deposits.
New York: John Wiley, pp. 71^136.
Campbell, I. H. & Turner, J. S. (1987). A laboratory investigation of
assimilation at the top of a basaltic magma chamber. Journal of
Geology 95, 155^172.
Camus, F. (1975). Geology of El Teniente orebody with emphasis on
wall-rock alteration. Economic Geology 70, 1341^1372.
Camus, F. (2003). Geolog|¤ a de los Sistemas Porf|¤ ricos en los Andes de Chile.
Santiago: Servicio Nacional de Geolog|¤ a y Miner|¤ a, 267 p.
Cannell, J. (2004). El Teniente porphyry copper^molybdenum deposit,
Central Chile. PhD thesis, University of Hobart, 317 p.
Cannell, J., Cooke, D. R., Walshe, J. L. & Stein, H. (2005). Geology,
mineralization, alteration, and structural evolution of the El
Teniente porphyry Cu^Mo deposit. Economic Geology 100, 979^1003.
Cannell, J., Cooke, D. R., Walshe, J. L. & Stein, H. (2007). Geology,
mineralization, alteration and structural evolution of the El
Teniente porphyry Cu^Mo depositçA reply. Economic Geology 102,
1171^1180.
Carroll, M. R. & Rutherford, M. J. (1985). Sulfide and sulfate saturation in hydrous silicate melts. Journal of Geophysical Research 90,
C601^C612.
Charrier, R., Baeza, O., Elgueta, S., Flynn, J. J., Gana, P., Kay, S. M.,
Mun‹oz, N., Wyss, A. R. & Zurita, E. (2002). Evidence for
Cenozoic extensional basin development and tectonic inversion
south of the flat-slab segment, southern Central Andes, Chile
(338^368S.L.). Journal of South American Earth Sciences 15, 117^139.
Charrier, R., Pinto, L. & Rodr|¤ guez, M. P. (2007).
Tectonostratigraphic evolution of the Andean orogen in Chile.
In: Moreno, T. & Gibbons, W. (eds) The Geology of Chile. London:
Geological Society, pp. 21^114.
Charrier, R., Far|¤ as, M. & Maksaev, V. (2009). Evolucio¤n tecto¤nica,
pelogeogra¤fica y metaloge¤nica durante el Cenozoico en los Andes
de Chile Norte y Central e implicaciones para las regiones adyacentes de Bolivia y Argentina. Revista de la Asociacio¤ n Geolo¤ gica
Argentina 65, 5^35.
Cherniak, D. J. & Watson, E. B. (2003). Diffusion in zircon.
In: Hanchar, J. M. & Hoskin, P. W. O. (eds) Zircon. Mineralogical
Society of America and Geochemical Society, Reviews in Mineralogy and
Geochemistry 53, 113^143.
Chiaradia, M., Merino, D. & Spikings, R. (2009). Rapid transition to
long-lived deep crustal magmatic maturation and the formation of
giant porphyry-related mineralization (Yanacocha, Peru). Earth
and Planetary Science Letters 288, 505^515.
Cline, J. S. & Bodnar, R. J. (1991). Can economic porphyry copper
mineralization be generated by a typical calc-alkaline melt?
Journal of Geophysical Research 96, 8113^8126.
Cloos, M. (2002). Bubbling magma chambers, cupolas and porphyry
copper deposits. In: Ernst, G. (ed.) Frontiers in Geochemistry: Organic,
Solution, and Ore Deposit Geochemistry. International Book Series 6,
191^217.
CODELCO. (2010). Annual Memoir. Santiago: CODELCO, 236 p.
Core, D. P., Stephen, E. K. & Essene, E. J. (2006). Unusually Cu-rich
magmas associated with giant porphyry deposits: evidences from
Bingham, Utah. Geology 34, 41^44.
Cornejo, P., Tosdal, R. M., Mpodozis, C., Tomlinson, A. J., Rivera, O.
& Fanning, C. M. (1997). El Salvador, Chile porphyry copper deposit revisited: geological and geochronologic framework.
International Geology Review 39, 22^54.
Cuadra, P. (1986). Geocronolog|¤ a K^Ar del yacimiento El Teniente y
a¤reas adyacentes. Revista Geolo¤ gica de Chile 27, 3^26.
Davidson, J. P., Turner, S. P., Handley, H., Macpherson, C. &
Dosseto, A. (2007). Amphibole ‘sponge’ in arc crust? Geology 35,
787^790.
Deckart, K., Clark, A. H., Aguilar, R., Vargas, R., Bertens, A.,
Mortensen, J. & Fanning, M. (2005). Magmatic and hydrothermal
chronology of the giant R|¤o Blanco porphyry copper deposit,
Central Chile: implications of an integrated U^Pb and 40Ar/39Ar
database. Economic Geology 100, 905^934.
Deckart, K., Godoy, E., Bertens, A., Jere¤z, D. & Saeed, A. (2010).
Barren Miocene granitoids in the central Andean metallogenic
belt, Chile: geochemistry and Nd^Hf and U^Pb isotope systematics. Andean Geology 37, 1^31.
DePaolo, D. J. (1981). Trace element and isotopic effects of combined
wall rock assimilation and fractional crystallization. Earth and
Planetary Science Letters 53, 189^202.
Eiler, J. M. (2001). Oxygen isotope variations in basaltic lavas and
upper mantle rocks. In: Valley, J. W. & Cole, D. (eds) Stable Isotope
Geochemistry. Mineralogical Society of America and Geochemical Society,
Reviews in Mineralogy and Geochemistry 43, 319^364.
Far|¤ as, M., Charrier, R., Carretier, S., Martinod, J., Fock, A.,
Campbell, D., Ca¤ceres, J. & Comte, D. (2008). Late Miocene high
1117
JOURNAL OF PETROLOGY
VOLUME 53
and rapid surface uplift and its erosional response in the Andes of
Central Chile (338^358S). Tectonics 27, TC1005.
Far|¤ as, M., Comte, D., Charrier, R., Martinod, J., Tassara, A. &
Fock, A. (2010). Crustal-scale structural architecture of the Central
Chile Andes based on 3D seismic tomography, seismicity, and surface geology: Implications for mountain building in subduction
zones. Tectonics 29, TC3006.
Faunes, A. (1981). Caracterizacio¤n de la mineralog|¤ a y de la alteracio¤n
en un sector del stock tonal|¤ tico del yacimiento El Teniente. BSc
thesis, University of Chile, Santiago, 175 p.
Ferry, J. M. & Watson, E. B. (2007). New thermodynamic models and
revised calibrations for the Ti-in-zircon and Zr-in-rutile thermometers. Contributions to Mineralogy and Petrology 154, 429^437.
Fock, A., Charrier, R., Far|¤ as, M. & Mun‹oz, M. (2006). Fallas de
vergencia oeste en la Cordillera Principal de Chile Central: inversio¤n de la cuenca de Abanico (338^348S). Revista de la Asociacio¤ n
Geolo¤ gica Argentina, Special Publication Series 6, 48^55.
Fu, B., Page, F. Z., Cavosie, A. J., Clechenki, C. C., Fournelle, J.,
Kita, N. T., Lackey, J. S., Wilde, S. A. & Valley, J. W. (2008).
Ti-in-zircon thermometry: applications and limitations.
Contributions to Mineralogy and Petrology 156, 197^212.
Garrido, C. J., Bodinier, J.-L., Burg, J.-P., Zeilinger, G., Hussain, S. S.,
Dawood, H., Chaudhry, M. N. & Gervilla, F. (2006). Petrogenesis
of mafic garnet granulite in the lower crust of the Kohistan
paleo-arc complex (Northern Pakistan): implications for intracrustal differentiation of island arcs and generation of continental
crust. Journal of Petrology 10, 1873^1914.
Getsinger, A., Rushmer, T., Jackson, M. D. & Baker, D. (2009).
Generating high Mg-numbers and chemical diversity in tonalite^
trondhjemite^granodiorite (TTG) magmas during melting and
melt segregation in the continental crust. Journal of Petrology 50,
1935^1954.
Giambiagi, L. B. & Ramos, V. (2002). Structural evolution of the
Andes between 33830 and 33845’S, above the transition zone between the flat and normal subduction segment, Argentina and
Chile. Journal of South American Earth Sciences 15, 99^114.
Gonza¤lez, R. (2006). Petrograf|¤ a, geoqu|¤ mica y microtermometr|¤ a
de los intrusivos fe¤lsicos del yacimiento El Teniente. BSc thesis,
University of Concepcio¤n, 149 p.
Goodge, J. W. & Vervoort, J. D. (2006). Origin of Mesoproterozoic
A-type granites in Laurentia: Hf isotope evidence. Earth and
Planetary Science Letters 243, 711^731.
Hedenquist, J. W. & Lowenstern, J. B. (1994). The role of magmas in
the formation of hydrothermal ore deposits. Nature 370, 519^527.
Hiess, J., Nutman, A. P., Bennett, V. C. & Holden, P. (2008).
Ti-in-zircon thermometry applied to contrasting Archean metamorphic and igneous systems. Chemical Geology 247, 323^338.
Hildreth, W. & Moorbath, S. (1988). Crustal contributions to arc magmatism in the Andes of Central Chile. Contributions to Mineralogy
and Petrology 98, 455^489.
Hitschfeld, M. (2006). Petrograf|¤ a y geoqu|¤ mica de los intrusivos leucocra¤ticos del sector sur este del yacimiento El Teniente, VI
Regio¤n, Chile. BSc thesis, University of Concepcio¤n, 129 p.
Hoskin, P. W. O. (1998). Minor and trace element analysis of natural
zircon (ZrSiO2) by SIMS and laser ablation ICPMS: a consideration and comparison of two broadly competitive techniques.
Journal of Trace and MicroprobeTechniques 16, 301^326.
Hoskin, P. W. O. & Schaltegger, U. (2003). The composition of zircon
and igneous and metamorphic petrogenesis. In: Hanchar, J. M. &
Hoskin, P. W. O. (eds) Zircon. Mineralogical Society of America and
Geochemical Society, Reviews in Mineralogy and Geochemistry 53, 27^62.
Howell, F. H. & Molloy, J. S. (1960). Geology of the Braden orebody,
Chile, South America. Economic Geology 55, 863^905.
NUMBER 6
JUNE 2012
Huppert, H. E. & Sparks, R. S. J. (1988). The generation of granitic
magmas by intrusion of basalt into continental crust. Journal of
Petrology 29, 599^624.
Ickert, R. B., Hiess, J., Williams, I. S., Holden, P., Ireland, T. R.,
Lanc, P., Schram, N., Foster, J. J. & Clement, S. W. (2008).
Determining high precision, in situ, oxygen isotope ratios with
SHRIMP II: analyses of MPI-DING silicate-glass reference materials and zircon from contrasting granites. Chemical Geology 257,
114^128.
Ireland, T. R. & Wlotzka, F. (1992). The oldest zircons in the solar
system. Earth and Planetary Science Letters 109, 1^10.
Isacks, B. L. (1988). Uplift of the Andean plateau and bending of the
Bolivian orocline. Journal of Geophysical Research 93, 3211^3231.
Jagoutz, O., Muntener, O., Ulmer, P., Pettke, T., Burg, J.-P.,
Dawood, H. & Hussain, S. (2007). Petrology and mineral chemistry
of lower crustal intrusions: the Chilas Complex, Kohistan (NW
Pakistan). Journal of Petrology 48, 1895^1953.
Kay, S. M. & Mpodozis, C. (2001). Central Andes ore deposits linked
to evolving shallow subduction systems and thickening crust. GSA
Today 11, 4^9.
Kay, S. M. & Mpodozis, C. (2002). Magmatism as a probe to the
Neogene shallowing of the Nazca plate beneath the modern
Chilean flat-slab. Journal of South American Earth Sciences 15, 39^57.
Kay, S. M., Godoy, E. & Kurtz, A. (2005). Episodic arc migration,
crustal thickening, subduction erosion, and magmatism in the
South^Central Andes. Geological Society of America Bulletin 117, 67^88.
Klemm, L. M., Pettke, T., Heinrich, C. A. & Campos, E. (2007).
Hydrothermal evolution of the El Teniente deposit, Chile: porphyry Cu^Mo ore deposition from low-salinity magmatic fluids.
Economic Geology 102, 1021^1045.
Kusakabe, M., Hori, M., Nakagawa, S., Matsushisa, Y., Ojeda, J. M.
& Serrano, L. M. (1984). Oxygen and sulfur isotopic compositions
of quartz, anhydrite and sulfide minerals from the El Teniente
and R|¤o Blanco porphyry copper deposits, Chile. Bulletin of the
Geological Survey of Japan 35, 583^614.
Kusakabe, M., Hori, M. & Matsushisa,Y. (1990). Primary mineralization^alteration of the El Teniente and R|¤o Blanco porphyry
copper deposits, Chile: stable isotopie, fluid inclusion and Mgþ2/
Feþ2/Feþ3 ratios of hydrothermal fluids. In: Herbert, H. K. &
Ho, S. E. (eds) Stable Isotopes and Fluid Processes in Mineralization.
Perth: University of Western Australia, pp. 244^259.
Laursen, J., Scholl, D. W. & Von Huene, R. (2002). Neotectonic deformation of the Central Chile margin: deep water forearc basin
formation in response to hot spot ridge and seamount subduction.
Tectonics 21, 1038, doi:10.1029/2001TC901023.
Lindgren, W. (1933). Mineral Deposits, 4th edn. New York: McGraw^
Hill, 930 p.
Lindgren, W. & Bastin, E. S. (1922). Geology of the Braden mine,
Rancagua, Chile. Economic Geology 17, 863^905.
Lo¤pez, S & Castro, A. (2001). Determination of the fluid-absent
solidus and supersolidus phase relationships of MORB-derived
amphibolites in the range 4^14 kbar. American Mineralogist 86,
1396^2001.
Ludwig, K. R. (2003). Isoplot 3.00. Berkeley Geochronology Center Special
Publication 4.
Maksaev, V. & Zentilli, M. (1988). Marco metaloge¤nico regional de los
megadepo¤sitos de tipo Po¤rfido Cupr|¤ fero del Norte Grande de
Chile. In: Proceedings V Congreso Geolo¤ gico Chileno Santiago 1:
B181^B212 (Acta).
Maksaev, V., Munizaga, F., McWilliams, M., Fanning, C. M.,
Mathur, R., Ruiz, J. & Zentilli, M. (2004). New chronology for El
Teniente, Chilean Andes, from U^Pb, 40Ar/39Ar, Re^Os, and
fission-track dating: implications for the evolution of a supergiant
1118
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
porphyry Cu^Mo deposit. Society of Economic Geologists Special
Publications 11, 15^54.
Maksaev, V., Townley, B., Palacios, C. & Camus, F. (2007). Metallic
ore deposits. In: Moreno, T. & Gibbons, W. (eds) The Geology of
Chile. London: Geological Society, pp. 180^199.
Maksaev, V., Munizaga, F., Zentilli, M. & Charrier, R. (2009). Fission
track thermochronology of Neogene plutons in the Principal
Andean Cordillera of Central Chile (33^358S): implications for
tectonic evolution and porphyry Cu^Mo mineralization. Revista
Geolo¤ gica de Chile 36, 153^171.
Mathur, R., Ru|¤ z, J. & Munizaga, F. (2000). Relationship between
copper tonnage of Chilean base-metal porphyry deposits and Os
isotope ratios. Geology 28, 555^558.
Mathur, R., Ru|¤ z, J. R. & Munizaga, F. M. (2001). Insights into
Andean metallogenesis from the perspective of Re^Os analyses of
sulfides. In: Extended Abstracts III South American Symposium on Isotope
Geology, Puco¤n, 500^503 (CD).
McDonough, D. F. & Sun, S.-s. (1995). The composition of the earth.
Chemical Geology 120, 223^253.
McKee, E. H. & Noble, D. C. (1989). Cenozoic tectonic events, magmatic pulses, and base- and precious-metal mineralization in the
central Andes. In: Ericksen, G. E., Can‹as, M. T. & Reinemund, J.
A. (eds) Geology of the Andes and its Relation to Hydrocarborn and
Mineral Resources. Earth Science Series, Vol. 11. Houston, TX:
Circum-Pacific Council for Energy and Mineral Resources,
pp. 189^194.
Montecinos, P., Scha«rer, U., Vergara, M. & Aguirre, L. (2008).
Lithospheric origin of Oligocene^Miocene magmatism in Central
Chile: U^Pb ages and Sr^Pb^Hf isotope composition of minerals.
Journal of Petrology 49, 555^580.
Mpodozis, C. & Ramos, V. A. (1989). The Andes of Chile and
Argentina. In: Ericksen, G. E., Can‹as, M. T. & Reinemund, J. A.
(eds) Geology of the Andes and its Relation to Hydrocarbon and Mineral
Resources. Earth Science Series, Vol. 11. Houston, TX: Circum-Pacific
Council for Energy and Mineral Resources, pp. 59^90.
Mungall, J. (2002). Roasting the mantle: slab melting and the genesis
of major Au and Au-rich Cu deposits. Geology 30, 915^918.
Munizaga, F., Maksaev, V., Fanning, C. M., Giglio, S., Yaxley, G. &
Tassinari, C. C. G. (2008). Late Paleozoic^Early Triassic magmatism on the western margin of Gondwana: Collahuasi area,
Northern Chile. Gondwana Research 13, 407^427.
Mun‹oz, M., Deckart, K., Charrier, R. & Fanning, C. M. (2009). New
geochronological data on Neogene^Quaternary intrusive rocks
from the high Andes of Central Chile (33815’^34800’S).
In: Proceedings XII Congreso Geolo¤ gico Chileno, Santiago, S8^008.
Oyarzu¤n, R., Ma¤rquez, A., Lillo, J., Lo¤pez, I. & Rivera, S. (2001).
Giant versus small porphyry copper deposits of Cenozoic age in
northern Chile: adakitic versus normal calc-alkaline magmatism.
Mineralium Deposita 36, 794^798.
Pardo-Casas, F. & Molnar, O. (1987). Relative motion of the Nazca
(Farallon) and South American plate since Late Cretaceous times.
Tectonics 6, 233^248.
Perello¤, J., Sillitoe, R. H., Brockway, H., Posso, H. & Mpodozis, C.
(2009). Contiguous porphyry Cu^Mo and Cu^Au mineralization
at Los Pelambres, central Chile. In: Proceedings XII Congreso
Geolo¤ gico Chileno, Santiago, S11^026.
Petford, N. & Gallagher, K. (2001). Partial melting of mafic (amphibolitic) lower crust by periodic influx of basaltic magma. Earth
and Planetary Science Letters 193, 483^499.
Pupin, J. P. (1980). Zircon and granite petrology. Contributions to
Mineralogy and Petrology 73, 207^220.
Rabbia, O. M. (2002). Cristaloqu|¤ mica de rutilo y anatasa en sistemas
de po¤rfidos cupr|¤ feros andinos: evaluacio¤n de su uso como
monitores de la actividad de metales en flu|¤ dos hidrotermales corticales. PhD thesis, University of Chile, Santiago, 147 p.
Rabbia, O. M., Reich, M., Herna¤ndez, L. B., King, R. & Lo¤pez
Escobar, L. (2000). High-Al TTG-like suite at the El Teniente porphyry copper deposit Chile1. In: Proceedings IX Congreso Geolo¤ gico
Chileno, Puerto Varas, 326^300.
Ramos, V., Zapata, T., Cristallini, E. & Introcaso, A. (2004). The
Andean thrust systemçlatitudinal variations in structural styles
and orogenic shortening. In: McClay, K. R. (ed.) Thrust Tectonics
and Hydrocarborn Systems. AAPG Memoirs 82, 30^50.
Rapp, R. P. & Watson, E. B. (1995). Dehydration melting of metabasalt at 8^32 kbar: implications for crust^mantle boundary recycling. Journal of Petrology 36, 891^893.
Reich, M. (2001). Estudio petrogra¤fico, mineraloqu|¤ mico y geoqu|¤ mico de los cuerpos intrusivos de Sewell y La Huifa, yacimiento
El Teniente, VI Regio¤n, Chile. BS. thesis, University of
Concepcio¤n, 104 p.
Richards, J. P. (2003). Tectono-magmatic precursors for porphyry
Cu^(Mo^Au) deposit formation. Economic Geology 98, 1515^1533.
Richards, J. P. (2005). Cumulative factors in the generation of giant
calc-alkaline porphyry Cu deposits. In: Porter, T. M. (ed.) Super
Porphyry Copper and Gold Deposits: A Global Perspective, Vol. 1. Adelaide:
PGC Publishing, pp. 7^25.
Richards, J. P. (2009). Postsubduction porphyry Cu^Au and epithermal Au deposits: products of remelting of subduction-modified
lithosphere. Geology 37, 247^250.
Richards, J. P., Boyce, A. J. & Pringle, M. S. (2001). Geologic evolution of the Escondida area, Northern Chile: a model for spatial an
temporal localization of porphyry Cu mineralization. Economic
Geology 96, 271^305.
Rivera, O. & Falco¤n, M. (2000). Las formaciones Farellones, Coya^
Machal|¤ y Abanico en los alrededores del yacimiento El Teniente:
secuencias de cuencas volcano-tecto¤nicas transversales del
Oligoceno^Mioceno de Chile central (33845’^34830’ LS).
Proceedings IX Congreso Geolo¤ gico Chileno, Puerto Varas 1, 819^823.
Rojas, A. (2003). Petrograf|¤ a y geoqu|¤ mica del Po¤rfido Teniente,
ubicado en el sector norte del yacimiento El Teniente, Provincia
de Cachapoal, VI Regio¤n, Chile. BSc thesis, University of
Concepcio¤n, 133 p.
Sen, C. & Dunn, T. (1994). Dehydration melting of a basaltic composition amphibolite at 1·5 and 2·0 GPa: implications for the origin of
adakites. Contributions to Mineralogy and Petrology 117, 394^409.
Shafiei, B., Haschke, M. & Shahabpour, J. (2009). Recycling of orogenic arc crust triggers porphyry Cu mineralization in Kerman
Cenozoic arc rocks, southeastern Iran. Mineralium Deposita 44,
265^283.
Sillitoe, R. H. (1972). A plate tectonic model for the origin of porphyry
copper deposits. Economic Geology 67, 184^197.
Skewes, A. & Holmgren, C. (1993). Solevantamiento andino, erosio¤n y
emplazamiento de brechas mineralizadas en el depo¤sito de cobre
porf|¤ dico Los Bronces, Chile central (338S): Aplicacio¤n de
geotermometr|¤ a de inclusiones fluidas. Revista Geolo¤ gica de Chile 20,
71^84.
Skewes, A. & Stern, C. R. (2007). Geology, mineralization, and structural evolution of the El Teniente porphyry Cu^Mo depositçA
discussion. Economic Geology 102, 1165^1180.
Skewes, A., Are¤valo, A., Holmgren, C. & Stern, C. R. (2001). Stable
isotope evidence for the formation from magmatic fluids of the
mineralized breccias in Los Bronces and El Teniente copper deposits, Central Chile. Extended Abstracts III South American
Symposium on Isotope Geology, Puco¤ n 531^534 (CD).
Skewes, A., Are¤valo, A., Floody, R. & Zun‹iga, P. (2002). The giant El
Teniente breccia deposit: hypogene copper distribution and
1119
JOURNAL OF PETROLOGY
VOLUME 53
emplacement. In: Goldfarb, R. J. & Nielson, R. L. (eds) Global
Exploration 2002: Integrated Methods of Discovery. Society of Economic
Geologists Special Publication 9, 299^332.
Skewes, A., Are¤valo, A., Floody, R., Zun‹iga, P. & Stern, C. R. (2005).
The El Teniente Megabreccia Deposit, the world’s largest copper
deposit. In: Porter, T. M. (ed.) Super Porphyry Copper and Gold
Deposits: A Global Perspective, Vol. 1. Adelaide: PGC Publishing,
pp. 88^113.
So«derlund, U., Patchett, P. J., Vervoort, J. D. & Isachsen, C. E. (2004).
The 176Lu decay constant determined by Lu^Hf and U^Pb isotope
systematics of Precambrian mafic intrusions. Earth and Planetary
Science Letters 219, 311^324.
Stern, C. R. (1991). Role of subduction erosion in the generation of
Andean magmas. Geology 19, 78^81.
Stern, C. R. & Skewes, A. (2005). Origin of giant Miocene and
Pliocene Cu^Mo deposits in Central Chile: Role of ridge subduction, decreased subduction angle, and long-lived, batholith size,
open-system magma chambers. In: Porter, T. M. (ed.) Super
Porphyry Copper and Gold Deposits: A Global Perspective, Vol. 1. Adelaide:
PGC Publishing, pp. 65^82.
Stern, C. R. & Skewes, M. A. (1995). Miocene to present magmatic
evolution at the northern end of the Andean Southern Volcanic
Zone, Central Chile. Revista Geolo¤ gica de Chile 22, 261^271.
Stern, C. R., Funk, J. A., Skewes, M. A. & Are¤valo, A. (2007).
Magmatic anhydrite in plutonic rocks at the El Teniente Cu^Mo
deposit, Chile, and the role of sulfur- and copper-rich magmas in
its formation. Economic Geology 102, 1335^1344.
Stern, C. R., Skewes, A. & Are¤valo, A. (2010). Magmatic evolution of
the giant El Teniente Cu^Mo deposit, Central Chile. Journal of
Petrology 52, 1591^1617, doi:10.1093/petrology/egq029.
Tassara, A., Go«tze, H.-J., Schmidt, S. T. & Hackney, R. (2006).
Three-dimensional density model of the Nazca plate and the
Andean continental margin. Journal of Geophysical Research 111,
B09404.
Tosdal, R. M. & Richards, J. P. (2001). Magmatic and structural controls on the development of porphyry Cu Mo Au deposits.
Reviews in Economic Geology 14, 157^181.
Valley, J. W., Chiarenzelli, J. R. & McLelland, J. M. (1994). Oxygen
isotope geochemistry of zircon. Earth and Planetary Science Letters
126, 187^206.
Valley, J. W., Kinny, P., Schulze, D. J. & Spicuzza, M. J. (1998).
Zircon megacrysts from kimberlite: oxygen isotope variability
among mantle melts. Contributions to Mineralogy and Petrology 133,
1^11.
van Dongen, M., Weinberg, R. F., Tomkins, A. G., Armstrong, R. A.
& Woodhead, J. D. (2010). Recycling of Proterozoic crust in
Pleistocene juvenile magma and rapid formation of the Ok
Tedi porphyry Cu^Au deposit, Papua New Guinea. Lithos 114,
282^292.
Vavra, G. (1994). Systematics of internal zircon morphology in major
Variscan granitoids types. Contributions to Mineralogy and Petrology
117, 331^334.
Vervoort, J. D. (2010). Hf analysis in zircon by LA-MC-ICPMS: promise and pitfalls. Geological Society of America, Abstracts with programs
42(5), 667.
Vervoort, J. D. & Blichert-Toft, J. (1999). Evolution of the depleted
mantle: Hf isotope evidence from juvenile rocks through time.
Geochimica et Cosmochimica Acta 63, 533^556.
Vry, V. H., Wilkinson, J. J., Seguel, J. & Milla¤n, J. (2010). Multistage
intrusion brecciation and veining at El Teniente, Chile: evolution
of a nested porphyry system. Economic Geology 105, 119^153.
Watson, E. B. & Cherniak, D. J. (1997). Oxygen diffusion in zircon.
Earth and Planetary Science Letters 148, 527^544.
NUMBER 6
JUNE 2012
Wolf, M. B. & Wyllie, P. J. (1993). Garnet growth during amphibolite
anatexisçimplications of a garnetiferous restite. Journal of Geology
101, 357^373.
Wolf, M. B. & Wyllie, P. J. (1994). Dehydration-melting of amphibolite
at 10 kbar: the effects of temperature and time. Contributions to
Mineralogy and Petrology 115, 369^383.
Woodhead, J. & Hergt, J. (2005). A preliminary appraisal of seven
natural zircon reference materials for in situ Hf isotope determination. Geostandards and Geoanalytical Research 29, 183^195.
Wyllie, P. J. & Wolf, M. B. (1993). Amphibolite dehydration-melting:
sorting out the solidus. In: Pritchard, H. M., Alabaster, T.,
Harris, N. B. W. & Neary, C. R. (eds) Magmatic Processes and Plate
Tectonics. Geological Society, London, Special Publications 76, 405^416.
Zhao, Z.-F. & Zheng, Y.-F. (2003). Calculation of oxygen isotope fractionation in magmatic rocks. Chemical Geology 193, 59^80.
Zheng, Y.-F. (1991). Calculation of oxygen isotope fractionation in
metal oxides. Geochimica et Cosmochimica Acta 55, 2299^2307.
Zheng, Y.-F. (1993a). Calculation of oxygen isotope fractionation
in anhydrous silicate minerals. Geochimica et Cosmochimica Acta 57,
1079^1091.
Zheng, Y.-F. (1993b). Calculation of oxygen isotope fractionation in
hydroxyl-bearing silicates. Earth and Planetary Science Letters 120,
247^263.
Zheng, Y.-F. & Fu, B. (1998). Estimation of oxygen diffusivity from
anion porosity in minerals. Geochemical Journal 32, 71^89.
A P P E N D I X : C A L C U L AT I O N O F
O I S O T O P E F R AC T I O N AT I O N
I N E X P E R I M E N TA L
D E H Y D R AT I O N M E LT I N G
R E AC T I O N S
The model presented in this study to estimate O isotope
fractionation in dehydration melting reactions (Fig. 12)
follows the calculation approach of Eiler (2001) using
the experimental results of Wolf & Wyllie (1994) and
Getsinger et al. (2009). According to this approach, the
melt is assumed to behave as the weighted sums of its normative constituents. This principle is applied in combination with fractionation factors to calculate the expected
isotopic distribution among all coexisting phases in the
observed assemblages in the melting experiments.
An initial composition of d18OWS ¼ 5·8ø has been
used for the whole system, which corresponds to that of
a mantle-derived basaltic magma (5·5ø) contaminated
with 3% of subducted sediment (15ø; Eiler, 2001). This
value has been assumed considering that a long-lived
deep MASH zone would ultimately have the O isotope signature of subduction-related, mantle-derived magmas.
The experimental results of Wolf & Wyllie (1994) and
Getsinger et al. (2009) reported for runs at various temperatures the volume per cent of melt and residue, the
melt major element chemical composition, and the
volume per cent of the minerals composing the residue.
For each experimental run, this information has been
used in the calculation of O isotope fractionation in the
system as described below, where calculations for the
1120
MUN‹OZ et al.
GENESIS OF PORPHYRY Cu DEPOSITS
8508C run of Wolf & Wyllie (1994) are shown as an example (Table A1).
(1) Separation of the melt into mineral components by
calculating the CIPW normative mineralogy (Table A1).
Melts from both experiments considered share most of the
normative constituents, which are quartz, albite, anorthite,
orthoclase, corundum, hypersthene, ilmenite and magnetite. Melts from the Getsinger et al. (2009) experiments also
incorporate apatite.
(2) Calculation of mineral-pair oxygen isotope fractionation factors. This has been carried out for a set of minerals
that include the melt normative constituents and the residue mineralogy. The fractionation factors for quartz^mineral pairs of Zheng (1991, 1993a, 1993b) have been used,
which are expressed in the form
(3) Calculation of oxygen isotope fractionation between
melt and minerals composing the residue. Considering
that 103ln (aMx1^Mx2) Mx1^Mx2, this can be done
assuming that the melt behaves as the sum of its normative
constituents (Table A1) and using the data from Table A3
(Table A4).
(4) Calculation of d 18OMelt according to the following
considerations. For the whole system
103 lnðaMx1Mx2 Þ ¼ ðA 106 Þ=T 2 þ ðB 103 Þ=T þ C
Table A2: Oxygen isotope fractionation factors for quartz^
mineral pairs
(with T in K). They include factors for anhydrous silicate
minerals, hydroxyl-bearing silicates and metal oxides
(Table A2). For each experimental run a matrix of oxygen
isotope fractionation factors has been calculated using
these data along with the corresponding temperature
(Table A3).
Table A1: Main parameters of the 8508C experimental run
(No. 150) of Wolf & Wyllie (1994) and melt normative
composition
Whole system
Phases and abundances* Chemical composition Normative composition
Melt
Grt
Opx
Hbl
Cpx
Pl
12·0
4·0
2·7
52·0
11·0
18·3
SiO2
TiO2
Al2O3
FeO
MnO
MgO
CaO
Na2O
K2O
wt %
Normative
minerals
Relative
proportion
64·79
0·43
19·07
4·51
0·18
1·51
7·95
1·25
0·29
Qtz
Ab
An
Kfs
Crs
Hy
Ilm
Mag
0·35
0·11
0·39
0·02
0·02
0·09
0·01
0·01
ðaÞ
where M and R correspond to the proportion of melt and
residue, respectively, and M þ R ¼ 1.
Mineral
A
B
C
Ref.*
Ab
Kfs
An
Opxy
Cpxz
Hbl
Py
Mag
Ilm
Crs
0·15
0·16
0·36
0·51
0·56
0·59
0·73
1·22
1·36
2·00
1·39
1·50
2·73
3·45
3·67
3·80
4·30
8·22
8·61
9·94
0·57
0·62
1·14
1·44
1·53
1·59
1·80
4·35
4·57
5·32
1
1
1
1
1
2
1
3
3
3
*Fractionation factors taken from: 1, Zheng (1993a); 2,
Zheng (1993b); 3, Zheng (1991).
yTaken after hypersthene.
zTaken after diopside.
Melt
vol. %
d18 OWS ¼ Md18 OMelt þ Rd18 OResidue
Table A3: Matrix of oxygen isotope fractionation factors
between quartz and selected minerals at 8508C
Qtz
Ab
Kfs
Mineral abbreviations used in this and the following tables:
Ab, albite; An, anorthite; Cpx, clinopyroxene; Crs, corundum; Grt, garnet; Hbl, hornblende; Hy, hypersthene; Ilm,
ilmenite; Kfs, K-feldspar; Mag, magnetite; Opx, orthopyroxene; Pl, plagioclase; Py, pyrope; Qtz, quartz.
*Modes indicated by Wolf & Wyllie (1994) are derived from
BSE and SEM-EDS analyses. In this work melt and garnet
abundances have been taken from the BSE data, when available, and remaining mineral abundances from the SEM-EDS
data recalculated for fitting the whole system at 100%.
An
Opx
Cpx
Hbl
Py
Mag
Ilm
1121
Ab
Kfs
An
Opx
Cpx
Hbl
Py
Mag
Il
Crs
0·79
0·84
1·58
2·04
2·18
2·26
2·61
3·94
4·17
5·12
0·06
0·79
1·25
1·40
1·47
1·82
3·15
3·39
4·33
0·73
1·19
1·34
1·42
1·76
3·09
3·33
4·27
0·46
0·61
0·69
1·03
2·36
2·60
3·54
0·15
0·23
0·57
1·90
2·14
3·08
0·08
0·43
1·75
1·99
2·93
0·35
1·67
1·91
2·85
1·33
1·57
2·51
0·24
1·18
0·94
JOURNAL OF PETROLOGY
VOLUME 53
Table A4: Oxygen isotope fractionation between melt and
minerals composing the residue of the 8508C experimental
run of Wolf & Wyllie (1994)
Melt
Mineral phases in the residue
Norm.
Hbl
An
Cpx
Opx
Py
NUMBER 6
JUNE 2012
Using (b) and considering that Melt^MxI ¼ d18OMelt d OMxI equation (a) can be rewritten as
X I d18 OMelt MeltMxI
d18 OWS ¼ Md18 OMelt þ
X X
d18 OWS ¼ M þ
I d18 OMelt IMeltMxI
ðcÞ
X
18
18
d OWS ¼ ðM þ RÞd OMelt IMeltMxI
X
d18 OMelt ¼ d18 OWS þ
IMeltMxI
18
min.
Qtz
0·78
0·55
0·76
0·70
0·90
Ab
0·16
0·08
0·15
0·13
0·19
An
0·27
0·00
0·24
0·18
0·41
Kfs
0·02
0·01
0·02
0·02
0·03
Crs
0·06
0·08
0·07
0·07
0·06
0·02
0·04
0·01
0·00
0·05
Ilm
0·02
0·02
0·02
0·02
0·01
Mag
0·02
0·03
0·03
0·03
0·02
0·46
1·07
0·92
1·50
Opx
Melt–Mx
1·15
Expression (c) allows calculation of the d18OMelt with
the assumed d18OWS ¼ 5·8ø and the data reported in
Table A4. Thus, a d18OMelt ¼ 6·7ø is obtained for the
8508C experimental run of Wolf & Wyllie (1994).
(5) Calculation of d18OResidue ¼ (d18OWS Md18OMelt)/
R, which is 5·7ø for the preceding example.
(6) Calculation of d 18O for zircon in equilibrium with
the melt (d18OZrc). This has been carried out according
to the relation given by Valley et al. (1994) where
d18ORock ¼ 0·06(wt % SiO2) 2·25 þ d18OZrc, using for
the d18ORock the d18OMelt. Thus a value of d 18OZrc ¼ 5·1ø
is obtained for the preceding example.
Rd18OResidue can be expressed in terms of the weighted
sum of the residue composing minerals as
X
Id18 OMxI
ðbÞ
with I being
P the corresponding proportion of the mineral
MxI and I ¼ R.
1122