©2015 Society of Economic Geologists, Inc.

Economic Geology, v. 110, pp. 241–251

SCIENTIFIC COMMUNICATIONS

ZIRCON COMPOSITIONAL EVIDENCE FOR SULFUR-DEGASSING FROM ORE-FORMING ARC MAGMAS*

John H. Dilles,1,† Adam J.R. Kent,1 Joseph L. Wooden,2 Richard M. Tosdal,3 Alison Koleszar,1
Robert G. Lee,1 and Lucian P. Farmer1
1College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, Oregon 97331-5501
2 School of Earth Sciences, Stanford University, Stanford, California 94305-2210
3 MDRU, Department of Earth and Ocean Sciences, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4

Abstract
Porphyry Cu (±Mo ±Au) and epithermal Au-Ag deposits are major sources of mined metals and are com-
monly formed by magmatic-hydrothermal fluids derived from hydrous magmas in Phanerozoic convergent
margin settings. The igneous rock assemblages associated with porphyry mineral deposits are common in mod-
ern convergent margin settings, but while many have produced acidic magmatic fluids, very few, past or present,
have produced sufficient metal, chlorine, and sulfur enrichments necessary to engender an ore deposit. The
reasons for this remain uncertain.
We report SHRIMP-RG ion microprobe analyses of hafnium, titanium and rare earth element (REE) abun-
dances in zircon, a nearly ubiquitous and robust trace mineral in crustal magmas. Comparison of the composi-
tions of zircons in ore-forming and barren granitic plutons indicate that ore-forming granites crystallized at
relatively low temperature and have relatively small negative europium anomalies (mostly EuN/EuN* ≥0.4).
We interpret these small zircon europium anomalies to indicate oxidizing magmatic conditions and hypothesize
that in many cases this reflects oxidation due to SO2 degassing from magmas with a relatively low Fe/S ratio.
Oxidation of europium and iron in the melt is produced by reduction of magmatic sulfate (S6+) to SO2 (S4+)
upon degassing. This interpretation reinforces the important role of oxidized sulfur-rich fluids in porphyry and
epithermal mineral deposit formation. Zircon compositions thus may be used to identify ancient magmas that
released significant amounts of SO2-rich gases, and regional surveys of zircon composition are potentially a valu-
able tool for mineral exploration.

Introduction calc-alkaline to mildly alkaline intermediate to silicic com-

The annual global consumption of copper, gold, silver, and positions, high water contents, and high oxidization state
molybdenum exceeds US$300 billion and will increase dra-
matically in the future. Porphyry Cu (±Mo ±Au; Seedorff et 0 Cu-(Mo-Au)
Cu-(Mo-Au) ores
ores Volcanics
al., 2005; Sillitoe, 2010) and epithermal Au-Ag (Simmons et al.,
2005) deposits are the most important sources of these metals
and predominantly occur in the circum-Pacific and Tethyan 10 H2O-SO2
degassing ~700°C
mountain belts. The igneous rock assemblages associated with of sulfate-rich
these occurrences are also common in modern convergent dacite magma
margin settings and many released acidic hydrothermal fluids, 20
but very few, past or present, have produced sufficient met- High-Al hbl ~900°C
als, chlorine, and sulfur necessary to engender an ore deposit crystallization &
km (depth)

sulfate saturation
(Dilles, 1987; Kay and Mpodozis, 2001; Rohrlach and Loucks, 30
2005; Chambefort et al., 2008; Richards, 2011a, b). Discovery
of new deposits is costly and difficult, and chemical finger-
MASH zone
printing of likely exploration targets remains a grail. 40 produces andesite
Ore-producing and other subduction-related arc magmas high in S, Cu, Cl,
are formed by complex processes that include hydrous mafic H 2O ~1000°C

magma input into MASH zones at the base of the crust, fol- base of crust
50
lowed by evolution and ascent of magma to the upper crust upper mantle hydrous
basalt
(Hildreth and Moorbath, 1988; Richards, 2003, 2011a, b;
Annen et al., 2006; Tosdal et al., 2009; Fig. 1). Most ore- Fig. 1. Cartoon crustal section showing injection of basalt from mantle
forming magmas display gross compositional characteristics and evolution of melts through a deep crustal MASH zone with amphibole,
pyroxene, and garnet fractionation, midcrustal chambers, and upper cham-
that are similar to nonmineralizing arc magmas, including bers where magmas crystallize to plutons and produce magmatic gas that
rises, cools to form hydrothermal fluid, and generates porphyry- and epith-
† Corresponding author: e-mail, dillesj@science.oregonstate.edu ermal-type ore deposits (after Hildreth and Moorbath, 1988; Annen et al.,
*A digital supplement to this paper is available at http://economicgeology. 2006; Tosdal et al., 2009; Richards, 2011b; Chambefort et al., 2013). Zircon
org/ and at http://econgeol.geoscienceworld.org/. crystallizes synchronously with vapor separation in shallow magmas.

Submitted: August 22, 2013

0361-0128/15/4283/241-11 241 Accepted: May 15, 2014
242 SCIENTIFIC COMMUNCIATIONS

(>NNO + 1; Seedorff et al., 2005). Many ore-forming magma is that large amounts of sulfur may be dissolved as sulfate at
suites are emplaced periodically over a relatively long interval high temperature or sequestered in crystalline anhydrite and
(~1−>7 m.y.) and crystallize in the upper 10 km of the crust can be released to ore fluids during cooling, crystallization,
as plutonic complexes; hydrothermal fluids that produce most and volatile evolution as SO2 gas (Streck and Dilles, 1998;
ores are associated with and likely directly separate from the Chambefort et al., 2008; Table 1). Furthermore, upon satura-
younger and more silica-rich magmas (Dilles, 1987; Longo tion with water-rich vapor, the anhydrite may break down to
et al., 2010). Magmatic volatile constituents such as Cl, F, form SO2 that enters the vapor because of the large vapor/
CO2, and S species allow the hydrothermal fluid to transport melt partition coefficient of 50 to 1,000 at ~800°C (Scaillet et
economic metals and fix them to form sulfide-rich ores (Pok- al., 1998; Keppler, 1999, 2010; Wallace and Edmonds, 2011).
rowski et al., 2008; Chambefort et al., 2013). Nonetheless, despite this framework, many details of these
Porphyry Cu deposits are sulfur rich and may contain more magmatic processes remain poorly known and are debatable.
than 1,000 million metric tons (Mt) of sulfur in rock sulfides One particular limit to traditional petrologic studies is the
and sulfates with relatively high total S/Cu ratios of ~100 widespread subsolidus recrystallization or hydrothermal alter-
(Gustafson, 1979; Table 1). Large volcanic emissions of SO2 ation of primary magmatic minerals that occurs throughout
(50 wt % S) from arc volcanoes are also observed, e.g., 3.7 Mt many ore-forming intrusions. In particular, primary magmatic
from Soufriere Hills from 1995 to 2009, 80 to 109 kt/yr from and hydrothermal anhydrite is readily dissolved and removed.
Bizymianny from 2007 to 2010, and 20 Mt from Pinatubo in Zircon is a geochemically robust mineral that records
1991 (Pallister et al., 1992, 1996; Carn and Prata, 2010; López magmatic chemical composition through the late-stage pro-
et al., 2013; Table 1). High amounts of volcanic sulfur emis- cesses of crystallization, volatile evolution, and hydrothermal
sions and sulfur fixed in magmatic-hydrothermal ore deposits alteration. Zircon is also a sensitive indicator of the magmatic
are both consistent with high sulfur contents of many oxidized oxidation state via its multivalent Ce and Eu contents. Rare
magmas where SO42– is the dominant sulfur species at oxygen earth elements (REE) in magmas commonly occur in the +3
fugacities >NNO + 1 (Jugo, 2009; Wallace and Edmonds, valence and partition strongly (heavy REE) to weakly (light
2011). Where anhydrite is present in oxidized magmas that REE) into the zircon structure as a function of ionic radius.
have recently erupted, such as Pinatubo, measured sulfur Relative to trivalent REE oxidized Ce4+ is preferentially incor-
emissions are high (Table 1). Direct observations of magmatic porated and reduced Eu2+ is excluded (Hoskin and Schalteg-
anhydrite are few in ore-forming magmas (see Chambefort et ger, 2003). Ballard et al. (2002) proposed that the magnitude
al., 2008), but high sulfate contents of apatite preserve a record of the positive Ce anomaly and negative Eu anomaly in zir-
of sulfur-rich conditions (>1,000 ppm) and anhydrite satura- con can record magmatic oxidation state, and this has been
tion in magmas (Streck and Dilles, 1998; Parat and Holtz, confirmed by experiments on high-temperature basaltic melts
2005; Van Hoose et al., 2013). Note that the estimates of sulfur (Burnham and Berry, 2012). Ballard et al. (2002) concluded
contents in anhydrite-bearing magmas shown in Table 1 are a that porphyry copper deposits are only associated with intru-
minimum based on the amount of sulfate dissolved in melt, sions with zircon having high calculated Ce4+/ Ce3+ ratios and
and that comparison of sulfur solubility at 900°C with that small negative Eu anomalies (EuN/EuN* >0.4), and suggested
at <800°C (Carol and Rutherford, 1987; Luhr, 1990; Baker this might be understood in terms of the interdependent rela-
and Rutherford, 1996) suggests that more sulfur is stored tionships between oxygen fugacity of magmas and melt sulfur
in crystalline anhydrite than in melt (e.g., Van Hoose et al., speciation. We note that there should be a direct link between
2013). Therefore, a key implication of the presence of anhy- oxidation state of REE (Ce and Eu) and speciation of sulfur
drite or sulfur-rich apatite in oxidized ore-forming magmas between reduced S2− and oxidized S6+. In order to further test

Table 1. Sulfur Content and Oxidation States of Arc and Ore-Forming Magmas

Magma vol Sulfur in Cu in

Location (km3) Oxidation state Anhydrite1 magma (wt %) S (Mt) ores (Mt)2 Reference

Pinatubo, Philippines     3.5 NNO +3 Y  0.25 3 10 4 -- Pallister et al.,1992, 1996

Yanacocha, Peru   90 NNO +1.5-2 Y >0.15 5 1,000 6 >5 Chambefort et al., 2008, 2013;
   Longo et al., 2010
Yerington, Nevada >70 NNO+2 I >0.07 7 1,000 6   8 Dilles, 1987; Streck and Dilles, 1998
   (Luhr Hill granite)
Butte, Montana ? NNO+2 I >0.02 7 1,000 6 35 Field et al., 2005; Houston and
  Dilles, 2013
El Salvador, Chile ? NNO +1.5-2 I >0.02 5 1,000 6 15 Gustafson and Hunt, 1975; Lee, 2008

1Y = observed, I = inferred on basis of S-content of apatite

2 Copper contained in known mineral deposits (in millions of metric tonnes)
3 Based on SO emissions including dissolved in magma and excess SO gas, 1991 eruptions
2 2
4 SO gas emissions of 20 Mt from 1991 eruptions
2
5,7 Based on sulfate solubility at 950° and 875°C, respectively
6 Estimated total sulfur deposited as sulfide and anhydrite in ores at Yanacocha and El Salvador from references; Yerington district based on a) alteration

volumes from Ann-Mason (Dilles and Einaudi, 1992: K-silicate ~ 16 Bt @ 0.5% S; sericite-chlorite ~ 20 Bt @ 1.5% S, b) same amount of S from Yerington
mine, Bear, and MacArthur deposits; and c) 100 Mt of S for advanced argillic zones in volcanic rocks (Lipske and Dilles, 2000); Butte based on pre-Main
stage 4.9 Bt @ 0.49% Cu Czehura, 2006) and 1% S; additional 20 Bt pre-Main stage @ 1% S; 3 Bt of gray-sericite @ 4% S; Main Stage 10 Mt Cu production
from rocks with S/Cu > 30
 SCIENTIFIC COMMUNCIATIONS 243

and improve this hypothesis, here we provide additional REE 10,000

compositional data on zircon and further assess the mecha- QMD zircon
nisms for links between magmatic oxidation and the role of 1,000 (Y767)
sulfur during magmatic degassing to form porphyry Cu ore

Chondrite-normalized abundance
fluids. Note that our analysis also assumes that the relatively
100
high magmatic water contents (>~3 wt %) typical of many arc GP zircon
magmas would suppress early plagioclase crystallization. Thus, (Y781)
any negative Eu anomaly in zircon will only reflect redox con- 10 QMD whole rock
ditions. In contrast, in relatively dry magmas where substan-
GP whole rock
tial early plagioclase fractionation has occurred, a negative Eu 1
anomaly produced by plagioclase crystallization will accentuate
any negative anomaly in zircon caused by reducing conditions.
0.1
In this study, we report trace element compositions of zircon
determined by high-precision SHRIMP-RG ion microprobe
from a series of well-studied igneous rocks using methods 0.01
fully described by Mazdab and Wooden (2006) and Claiborne La Ce (Pr) Nd Sm Eu Gd Tb Dy Y Er Yb Lu
et al. (2010). The U/Pb isotope ages were determined simulta- Fig. 2. Whole-rock REE patterns and fields from the Yerington batho-
neously; they are not reported here but were used to exclude lith, Nevada, of zircon REE from early McLeod Hill quartz monzodiorite
inherited zircons. Samples include early barren to poorly min- (QMD, Y-767) and a mineralizing granite porphyry (GP, Y-781) derived from
eralized intrusions or volcanics and late-stage granodiorite to the Luhr Hill granite (samples from Dilles, 1987). QMD illustrates typical
negative EuN/EuN* anomaly on a nonmineralizing arc granite, whereas GP
granite porphyry intrusions directly associated with ores: El has a smaller negative EuN/EuN* anomaly typical of mineralizing intrusions.
Salvador, Chile (Gustafson and Hunt, 1975; Gustafson et al.,
2001; Lee, 2008); Yanacocha, Peru (Chambefort et al., 2008;
Longo et al., 2010); Yerington, Nevada (Dilles, 1987); Car- saturated and following prior studies are assigned an activity
lin (Muntean et al., 2011; Farmer, 2013), and Battle Moun- of TiO2 of 0.7. Variation of the TiO2 activity by 0.2 or variation
tain, Nevada (Farmer, 2013); and Butte, Montana (Lund et of titanium content in zircon as a function of sector zoning
al., 2002). Further details of these deposits and details of the (Lee, 2008) each produce errors of about 15° to 20°C, so the
analytical procedures used are given in supplementary data. calculated model temperatures vary within ±30°C (2σ). As
Comparable data are also available for porphyry deposits at documented in other studies of felsic melts, the hafnium con-
El Abra and Chuquicamata, Chile (Ballard et al., 2002); El tent and Hf/Zr ratio of zircon increase as the titanium content
Teniente, Chile (Muñoz et al., 2012); Oyu Tolgoi, Mongolia and Ti-in-zircon model crystallization temperature decrease
(Wainwright et al., 2011); Youlong, Tibet (Liang et al., 2006); (Claiborne et al., 2010). Despite some scatter most likely due
and central-eastern China (Wang et al., 2013), as well as the to analytical uncertainties and small variations in the activity
Gidginbung high sulfidation Au-Ag-(Cu) deposit, SE Austra- of TiO2 in melt, Figure 3 illustrates that the Hf content in
lia (Fu et al., 2009). zircon increases smoothly with decrease in Ti-in-zircon tem-
perature for both ore- and nonore-bearing samples. Zircon in
REE Patterns of Zircon samples of andesitic composition (Yanacocha CNN-1, CB-65,
Zircons in arc magma intrusions typically have low light REE and CNN-1; Yerington Y-767) have model temperatures of
and elevated heavy REE contents with distinctive positive 850° to 700°C. In contrast, zircon in all the granites, granite
Ce and negative Eu anomalies. In mineralized intrusions, the porphyries, and dacites associated with mineralization (except
Ce anomaly is typically larger and the Eu anomaly is typically Carlin Bu051c) yield lower model temperatures of 750° to
smaller in magnitude compared to nonmineralized intrusions 650°C consistent with zircon crystallization in near-eutectic
(Ballard et al., 2002). Figure 2 illustrates these features with the conditions close to the solidus of hydrous granite.
whole-rock and zircon REE patterns of samples from the Yer-
ington batholith, Nevada, including the early nonmineralized Europium Anomalies
McLeod Hill quartz monzodiorite (Y-767) and a mineralizing Zircons in the mineralizing intrusions typically have differ-
granite porphyry derived from the Luhr Hill granite (Y-781). ent europium anomalies compared to zircons in associated
nonmineralizing subduction-related intermediate to silicic
Ti-in-Zircon Temperature magmas. In most nonmineralizing granitoid suites, the chon-
The temperature of zircon crystallization can be estimated drite-normalized negative europium anomaly (EuN/EuN,*
from the titanium content if the TiO2 and SiO2 activities in where EuN* = (SmN*GdN)1/2) of zircon becomes more pro-
the melt are constrained (Watson and Harrison, 2005; Ferry nounced with increased Hf content, a proxy for decrease in
and Watson, 2007). The experimental calibration is for a pres- model temperature (Fig. 4). Figure 4 illustrates this pattern
sure of 1 GPa (10 kbars), and although Ti-in-zircon is mod- for several nonmineralized intrusions that predate mineraliza-
erately sensitive to pressure (Ferris et al., 2008), the zircons tion shortly (<1−10 m.y.) in locations such as the Los Picos-
in this study likely crystallized in the limited pressure range Pajonal, Chile; the Boulder batholith, Montana; the McLeod
observed for porphyry copper granitoids (1−3 kbars, e.g., Hill and Bear intrusions, Yerington. Additionally, regionally
Seedorff et al., 2005). Therefore, we neglect pressure effects extensive barren intrusions, including plutons in Nevada and
and use temperatures largely for comparison between dif- the Mojave desert, are part of the same Middle Jurassic arc
ferent samples. The magmas we have investigated are SiO2 suite that includes the Yerington batholith. The Cenozoic
244 SCIENTIFIC COMMUNCIATIONS

920
Associated Cu/Au ores
2 std dev
errors Yanacocha Volc, Peru Age (Ma)
880
DNS-1 8.4
YN-1A
VC-4 11.5
840
Titanium Temperature (°C)

CHQS-2 12.0
CB-65
800 DM014 12.4
CNN-1 14.4
FRAIL-2 15.5
760 Eocene dikes, Nevada
Carlin (Bu051c) 37
Battle Mt (LFBM3) 37
720 Yerington batholith, Nevada
near eutectic QMD ppy (LH-6) 165
temperatures Gran ppy (Y-781) 168.5
680 QMD (Y-767) 169.5
El Salvador, Chile
Latite dikes 41
640 L porphyry 42
A porphyry 42
K porphyry 42.5
600
7000 9000 11000 13000 15000 17000
Hf (ppm)
Fig. 3. Ti-in-zircon temperatures assuming a melt activity of TiO2 = 0.7 calculated after Ferry and Watson (2007) for
Yanacocha, Carlin, Yerington, and El Salvador magmas with SHRIMP-RG 206Pb/238U ages. Most mineralizing and dacite
composition igneous rocks (except the Carlin dike) yield temperatures of 750° to 650°C, whereas precursor andesite composi-
tion rocks yield higher temperatures (inset CL image of zircon).

Spirit Mountain granite pluton is a barren pluton associated petrogenesis. Most researchers have argued that fluid release
with bimodal extension-related magmatism. from subducted materials provides water, sulfur, chlorine, alka-
In contrast, in all these cases mineralizing intrusions have lis, and other components that flux the overlying mantle wedge
zircons with smaller negative EuN/EuN* anomalies compared and leads to melting and generation of deep oxidized mafic
to nonmineralized intrusions of similar age, geographic posi- precursors to erupted arc magmas (e.g., Richards, 2003, 2011b;
tion, and tectonic setting. The mineralized EuN/EuN* versus Kelley and Cottrell, 2009), whereas direct partial melting of
Hf trends also differ markedly between deposits. For exam- the subducted oceanic slab is considered an anomalous condi-
ple, EuN/EuN* is as low as 0.2 at Butte and El Abra, but EuN/ tion by most workers (Defant and Drummond, 1990: Mungall,
EuN* is more than 0.5 at El Salvador, Yanacocha, and Yering- 2002; Sun et al., 2012). Furthermore, in the magmatic models,
ton. As Hf increases, the zircon EuN/EuN* decreases slightly these hydrous basalts evolve in the lower crust to form hydrous,
at El Salvador, changes little at Butte, increases slightly at El oxidized, and sulfur-rich arc andesites and dacites character-
Abra, Chuquicamata, and Yanacocha, and increases more at ized by high Sr/Y via reaction with amphibolitic lower crust
Yerington (Fig. 4). In all cases in mineralized intrusions EuN/ or fractionation of amphibole ± garnet in the lower or mid-
EuN* has a large range at a given Hf content, in contrast to dle crust (cf. Kay and Mpodozis, 2001; Rohrlach and Loucks,
nonmineralized intrusions, such as the McLeod Hill and Bear 2005; Chiaradia et al., 2009; Lee et al., 2012; Loucks, 2014).
from Yerington that display a limited range and smooth varia- Such crustal processes most likely modify magma oxidation
tion of EuN/EuN* versus Hf. It may be noted that whole rocks state (Loucks, 2014). In plutonic suites related to porphyry Cu
from the Yerington batholith have no Eu anomaly (EuN/EuN* deposits, temporal evolution to late and more silicic magmas
≈ 1), including the nonmineralized McLeod Hill and Bear associated with ores is commonly also associated with increase
and mineralized Luhr Hill and porphyries (Fig. 1), so varia- in fO2 (Burnham and Ohmoto, 1980; Dilles, 1987), suggesting
tion in the zircon EuN/EuN* anomaly in different rocks is not that ore-forming magmas are further oxidized as part of volatile
correlated with the whole-rock Eu anomaly. loss to produce ore fluids. The sulfur-rich 1991 Pinatubo erup-
tions also provide direct evidence for shallow oxidation from
Discussion: Oxidation from Magmatic Degassing mafic magmas (NNO + 2) to dacites (NNO + 3) in a single
Many hydrous (~3−5 wt % water) arc magmas have elevated eruptive unit (Pallister et al., 1996). Metal sources for the ores
oxidation states (high fO2) in the range of NNO + 1 to NNO remain debatable because the evolved magmas associated with
+ 3 (Annen et al., 2006; Richards, 2011a), with mineralizing ores may have received sulfur, copper, and gold transferred
magmas commonly at the upper end of this range (Table 1). from mafic magma and other metals from crustal sources.
Although the origin of strong oxidation is debated, there is a The cause of late-magmatic oxidation has remained a con-
broad consensus that oxidization is related to subduction zone troversy in porphyry copper magmas (Burnham and Ohmoto,
 SCIENTIFIC COMMUNCIATIONS 245

0.9 1.0
Chuqui East ppy Yerington Batholith, NV

El Abra-Fortuna-Chqui
Opache ppy

Pre-Min Mineralized
Chuquicamata-El Abra, Chile
0.8 El Abra ppy Mineral & LH6 Late QMD ppy
South Gd Post-Mineral Y781 Ann-Mason ppy
0.7 Abra Ap, Gd, Di 0.8 99016 Buckskin ppy
Y750 Luhr Hill Gran
Los Picos Di
0.6 Y800 Bear Qz Monz
Pajonal Di
EuN/EuN*

Y767 McLeod
Jurassic, Hill QMD
Nevada

EuN/EuN*
0.5 0.6 other Juras plutons, NV

0.4 Mineralized

0.3 Pre-Mineral 0.4

Pre-Mineral
cogenetic (Hbl-Bio)
0.2

0.1 Pre-Mineral (Px-Bio) 0.2

0.0
A Hf (ppm) inherited zircons
6000 8000 10000 12000 14000
0.0
D Hf (ppm)

0.9 6000 8000 10000 12000 14000 16000 18000

Boulder Batholith, Montana Butte Modoc
0.8 Bou porphyry
lder DNS-1 Chaupi dacite, 8.6 Ma CB-65 L San Jose Ign 11.5 Ma
Bat Hell Canyon Pluton Yanacocha Volcs, Peru
holi YN-1A Yana dacite, 9.9 Ma CHQS-2 Azufre andesite, 12.1 Ma
th ( Butte Granite
0.7 Hell 1.0 DM-008 Encajon dac, 10.7 Ma DM-014 Quilish dac, 12.4/14.5 Ma
FRAIL-2, Co Frailes Ign, 15.5 Ma
Cyn Boulder Batholith DM-019 El Tapado dac, 11.1 Ma CNN-1, Atazaico And, 14.5 Ma
Plu
ton) VC-4 L San Jose Ign 11.5 Ma
EuN/EuN*

0.6 Boulder
Batholith
0.5 (Main Trend) 0.8
Mineralized Intrusions
0.4
EuN/EuN*

Mineralized 0.6
0.3
Sto
dd
ard
0.2 Rid
ge

0.1 0.4
Tu

B
Ra
Hf (ppm)
ff

ttle
sn
Bu ak
0.0 llio
nM
eM
6000 8000 10000 12000 14000 Tu tG
rtle tG ra
ran no
0.2 M
tT od
ior
dio
uf ite rite
f
0.9 DM014 DM019 DM008 CHQS-2 Spirit Mt.
YN-1A VC-4Pluton
DNS-1

0.8
El Salvador, Chile
0.0
E Hf (ppm)
Late, well-mineralized trend 6000 8000 10000 12000 14000 16000 18000
0.7
EuN/EuN*

0.6
El Salvador zircons
0.5
Age < 41.5 Ma
0.4 Age 41.5 - 42.5 Ma
Age 42.5 - 43.5 Ma
0.3
Age > 43.5 Ma
0.2 Early, poorly mineralized trend

0.1

0.0
C Hf (ppm)

6000 8000 10000 12000 14000

Fig. 4. Zircon EuN/EuN* vs. Hf for (A) El Abra and Chuquicamata, Chile, (B) Butte, Montana, (C) El Salvador, Chile,
(D) Yerington, Nevada, and (E) Yanacocha with comparison of mineralized with nonmineralized (gray shade) intrusions that
include Jurassic Nevada (J.E. Wright, S. Wyld, and J.L.Wooden, unpub. data), Boulder batholith (Wooden et al., 2008, and
unpub. data), Jurassic Mojave (Fohey-Breting et al., 2010), and Spirit Mountain batholith (Claiborne et al., 2010). Hf content
of zircon is a proxy for magmatic evolution and cooling. Note that for mineralizing igneous rocks, the EuN/EuN* values are
relatively constant with magmatic evolution.

1980). It is well established that during ascent of magmas H2 gas loss is negligible (Burgisser and Scaillet, 2007). Bell
from depth (<10 km or 3 kbars) to surface, degassing during and Simon (2011) did experiments and proposed that at high
depressurization releases water and substantial amounts of H2 volatile/melt ratios ranging from 1 to 3, loss of chloride-rich
gas that can produce oxidation of melt by 1 to 3 log units of vapors that preferentially transport Fe2+ relative to Fe3+ can
oxygen fugacity (Candela, 1986; Burgisser and Scaillet, 2007; cause increase of oxygen fugacity by 1 to 2 log units. Although
Fiege et al., 2014). Nonetheless, during degassing at constant attractive, this mechanism requires volatile/melt ratios or
pressure (1−3 kbars) consistent with pluton crystallization in salinities that are too great for most porphyry copper plutons.
the porphyry copper environment (Dilles, 1987), oxidation via For example, Audétat et al. (2008) summarized the types of
246 SCIENTIFIC COMMUNCIATIONS

fluids that coexist with mineralized magmas and found that magmas of central-eastern China (Wang et al., 2013), as well
some shallow magmas (<4-km depth) coexist with brine and as the magmas we have studied. For these reasons, improved
vapor, but found most magmas (depth >4-km fluid release: understanding of EuN/EuN* anomalies appears merited.
Seedorff et al., 2005) contain a single-phase fluid with rela- A negative EuN/EuN* anomaly in zircon is generally the
tively low amounts of dissolved solids (typically 3−7  wt % result of crystallization of plagioclase rich in Eu2+ prior to
NaCl equiv). Volatile/melt ratios are commonly low, e.g., the zircon. Common arc magmas are oxidized with log fO2 buff-
~80-km3 mineralizing Luhr Hill granite at Yerington most ered near NNO + 1 to >NNO + 2 and contain sufficient Eu2+
likely evolved about 5 wt % dissolved water (Dilles, 1987), and (~30−50%) to produce such negative EuN/EuN* anomalies
even if the magmas contained an extra 5 wt % of water as a in zircon (Wilke and Behrens, 1999). Therefore, the smaller
separate volatile phase (see Burgisser and Scaillet, 2007), the EuN/EuN* anomalies observed in zircons from mineralized
resultant maximum volatile/magma ratio of 1/10 would only intrusions compared to nonmineralized intrusions could
increase the oxygen fugacity of the melt by ca. 0.1 log unit via potentially result from suppression of plagioclase crystalliza-
the mechanism of Bell and Simon (2011). Below, we argue tion at high pressure and water content and would primarily
that degassing of SO2-rich volatiles is a likely mechanism to be a reflection of a necessary condition that ore-forming mag-
produce the oxidation observed. mas must also be hydrous (Richards, 2011a).
Ballard et al. (2002) demonstrated that both Ce4+/Ce3+ and However, it is also unlikely that differences we observe in
Eu3+/Eu2+ ratios are sensitive to the oxidation state of melt, zircon EuN/EuN* are related solely to plagioclase suppres-
and that high Ce4+/Ce3+ and small negative EuN/EuN* anom- sion and differences in water contents. For a start the non-
alies are characteristic of mineralizing arc magmas in the mineralized magmas we have studied are also oxidized and
Chuquicamata district, Chile. Ballard et al. (2002) empha- water rich (>3.5 wt % on the basis of hornblende present) and
sized calculated Ce4+/Ce3+ (CeIV/CeIII) ratios for estimation thus would have also experienced some early suppression of
of magmatic redox conditions. In the present study, we have plagioclase crystallization. Furthermore, almost all interme-
focused on the EuN/EuN* ratio, which is easily calculated diate and evolved arc magmas, mineralized and nonmineral-
because Sm, Eu, and Gd are relatively abundant in zircon ized, crystallize significant plagioclase, and it is the dominant
(0.5−20, 0.1−10, 1−100 ppm, respectively) and can be accu- modal phase in all the samples we have studied herein. Thus,
rately measured via laser ablation-ICP-MS or SHRIMP-RG. even where water contents suppress plagioclase crystallization
Although Ce4+/Ce3+ ratios of zircon are excellent indicators at higher temperatures by the time magmas have cooled to
of oxidation state of the melt as evidenced by positive CeN/ nearer to solidus temperatures (where zircon crystallizes, Fig.
CeN* anomalies in zircon, a number of issues make it dif- 3) significant plagioclase has most likely formed in both min-
ficult to estimate Ce4+/Ce3+. First, spectroscopic measure- eralized and nonmineralized magmas. This is supported by
ments of Ce4+/ Ce3+ have so far been unsuccessful because the high plagioclase modal abundances (20−60 vol %) in both
the ratio of ~10−3 is so low. Second, measurement of the the mineralized and nonmineralized plutons we have studied
magnitude of the CeN/CeN* = (CeN/(LaN*PrN)1/2 anomaly herein. Finally, even if there are distinct differences in the
is difficult because although Ce is abundant in zircon (5−10 temperature at which the onset of plagioclase crystallization
ppm), both La and Pr are very low in abundance (<1−200 occurs in mineralized versus nonmineralized plutons this is
and 10−500  ppb, respectively) and are susceptible to con- less likely to modify the EuN/EuN* of the residual liquid as
tamination by small melt and apatite inclusions that are com- Eu is less compatible in early anorthitic plagioclase relative
mon in zircon (Ballard et al., 2002; Lee, 2008; Wang et al., to the characteristic and more albitic plagioclase that forms
2013). In our studies using the SHRIMP-RG, the ion beam closer to the solidus in the mineralized intrusions. For exam-
only sputters to a depth of 1 to 2 mm in zircon and phospho- ple, Yerington porphyry plagioclase phenocrysts are unzoned
rous is monitored to avoid apatite inclusions, but nonethe- and sodic (24−26 mol % An, Dilles, 1987). From this we argue
less La contents are elevated above expected values (~1−25 that differences in water content in mineralized versus non-
ppb). To avoid these issues, Ballard et al. (2002) used an indi- mineralized magma compositions alone do not seem adequate
rect method to calculate Ce4+/Ce3+ using an estimate of Ce* to produce the observed EuN/EuN* anomalies in zircon for
from the linear regressions of REE concentration versus ion mineralized porphyries.
radius, and the assumption that whole-rock Ce concentration We prefer an alternative explanation that the small and
is the same as melt Ce concentration when the zircon crys- variable negative EuN/EuN* of zircon characteristic of ore-
tallized. Both these estimates introduce errors, in particu- forming magmas is the result strongly oxidized (≥ NNO +
lar because the Ti-in-zircon temperature of 750° to 650°C 1.5) conditions that become more oxidized during degassing
(Fig. 3) requires that zircon crystallizes late during magma of ore fluids. Strongly oxidized magmas convert Eu2+ to Eu3+
solidification after many REE-rich minerals (e.g., apatite, and therefore are Eu2+ poor, so that Eu is not incorporated
titanite, as well as plagioclase and hornblende) have crystal- into plagioclase but is incorporated in zircon as Eu3+. Further-
lized so that whole-rock Ce is not likely a good proxy for the more, strong oxidation is consistent with the requirement that
Ce concentration of melt. Wang et al. (2013) noted that in abundant oxidized magmatic sulfur allows metals to dissolve
their data set Ce4+/Ce3+ calculated by the method of Ballard in melt rather than form Cu-Au-rich magmatic sulfides (Can-
et al. (2002) does not discriminate for oxidation as well as dela, 1986; Candela and Holland, 1986; Halter et al., 2004;
EuN/EuN*. The recent calibration of the Ce4+/Ce3+ of zircon Brennecka, 2006) and also provides sulfur to ore fluids (Streck
as a function of magmatic temperature and oxygen fugacity and Dilles, 1998; Field et al., 2005; Chambefort et al., 2008).
(Trail et al., 2011) yields unrealistically high oxygen fugaci- Despite the fact that the majority of arc magmas are rela-
ties for silicic magmas including the Bishop Tuff, ore-forming tively water rich and oxidizing, water-content and oxidation
 SCIENTIFIC COMMUNCIATIONS 247

state commonly increase within a magmatic suite from early and Behrens (1999; Fig. 5A). Changes in EuN/EuN* with mag-
nonmineralized to late mineralized intrusions (Dilles, 1987; matic evolution will also be sensitive to differences in Fe/S of
Richards et al., 2012). the host magmas and in the resulting power of the magma
We argue that Eu anomalies of zircon in mineralizing intru- to buffer increases in fO2 related to SO2 degassing. Reduction
sions also reflect degassing of SO2-rich magmatic volatiles of sulfate to SO2 via reaction (2) produces oxidation of Fe2+
from sulfate-rich melts and variations in magmatic Fe/S ratios, to Fe3+ that varies as a function of the Fe/S mass ratio of the
which affect the ability of the magma to buffer this oxidation. melt (Fig. 5A). The degree of oxidation of Fe2+ to Fe3+ and
Porphyry copper magmas are both oxidized and very sulfur Eu2+ to Eu3+ can be modeled using simple stepwise crystal-
rich (Gustafson, 1979; Einaudi et al., 2003; Seedorff et al., lization and partitioning theory between crystals, melt, and
2005; Table 1). During cooling, oxidized and sulfur-rich sili- vapor previously developed by Candela and Holland (1986),
cate magma may saturate in anhydrite (Streck and Dilles, 1998; Candela (1989), Cline and Bodnar (1991), Cline (1995), and
Chambefort et al., 2008). Further cooling causes saturation in a Candela and Piccoli (1995). For an oxidized and sulfate-rich
water-rich fluid phase, and crystalline (xls) anhydrite may break high-temperature melt, prior to vapor saturation, cooling and
down to release CaO to the silicate melt (m) and SO2 to the crystallization of mafic minerals reduces the Fe content of the
vapor phase (v) via the reaction (Baker and Rutherford, 1996). melt, whereas oxidized sulfur may be sequestered in crystal-
line anhydrite. With further crystallization, the melt saturates
CaS6+O4 (xls) → CaO(m) + S4+O2(v) + 0.5 O2. (1)
with hydrous vapor, and anhydrite may break down to SO2 gas
Baker and Rutherford (1996) argued that anhydrite should that partitions strongly into the vapor via reaction (2). Figure
not readily break down in magmas, but anhydrite breakdown 5A illustrates for water-rich and oxidized magma the general
is predicted by the large partition coefficient between SO2 in effect of initial melt Fe/S ratios given typical partition coef-
vapor and SO42– in melt (~500−1,000 at NNO+2 and 800°C; ficients for crystal, melt, and vapor for Fe and S species. The
Scaillet et al., 1998). At strongly oxidized magmatic conditions model employed is very general because the partition coef-
of NNO + 2 and 700° to 800°C, the sulfur species in the vapor ficients and degree of crystallization required to produce
contains more SO2 than H2S (Whitney, 1988; Field et al., 2005; vapor saturation vary as a function of melt composition, melt
Zajacz et al., 2012). We note that our calculations are based volatile content, pressure, and temperature. Nonetheless, the
solely on SO2 as the sulfur species in the volatile phase, and models predict the general behavior because the underlying
that reduction of SO42– in the melt to H2S would produce more principals are well founded based on experimental data and
oxidation than in our calculations. Moreover, in our observa- modeling, and appropriate water contents and partition coef-
tions, anhydrite phenocrysts from the 1981 Pinatubo erup- ficients are used (see Candela and Piccoli, 1995, and refer-
tion are anhedral and resorbed, consistent with breakdown. ences therein).
Although sparse magmatic anhydrite has been observed in We have also tested whether fractionation of REE, in par-
some porphyry Cu intrusions (Deen et al., 1994; Audétat et ticular the preferential loss of Eu2+ compared to Eu3+ during
al., 2004; Chambefort et al., 2008), in most porphyry intru- magmatic degassing, could produce the observed varation in
sions anhydrite has likely quantitatively broken down via the zircon EuN/EuN*. Reed et al. (2000) determined partition
above reaction to supply sulfur to magmatic-hydrothermal coefficients at 800°C and 2 kbars (200 MPa) for magmatic
fluids (e.g., Chambefort et al., 2008). The reduction of sulfate vapor with a molality of Cl− =1.1 and melt, and found that
to SO2 requires oxidation of ferrous Fe in the melt or in Fe- DEu2+v/m = 0.04 compared to DSmv/m = 0.02 and DGdv/m = 0.02.
bearing phases via reactions such as the following: Using these values and a maximum vapor/melt ratio of 1/10,
degassing is predicted to increase EuN/EuN* very slightly
CaS6+O4 + 2 Fe2+O = CaO + S4+O2+ 2 Fe3+O1.5, (2)
(<0.02) and much less than we observe.
and In most cases of observed porphyry dikes associated with
porphyry Cu ores, the melts were crystal rich (~50 vol %)
3Fe2+S6+O4(m) → 3S4+O2 + O2 + Fe2+Fe3+2O4, (3)
upon emplacement and release of magmatic-hydrothermal
where reaction (2) is appropriate for anhydrite-bearing melt ore fluids, and they vary from andesite to rhyolite composition
and reaction (3) for sulfate-rich melts (Jugo, 2009; Jugo et al., (see review of Seedorff et al., 2005). Dacite (granodiorite) is
2005). Because of the reaction stoichiometry in reaction (2) the most common bulk composition, and the silicic melt at
and mass difference between S (32.06) and Fe (55.85), 3.5 high degrees of crystallization is commonly Fe poor (e.g., ~0.5
mass units of Fe are oxidized for each mass unit of S that is wt % Fe on the basis of 70−75 wt % silica melt inclusions at
reduced. Increase in the oxidation state of the magmas via Bajo Alumbrera; Halter et al., 2004). At low Fe/S mass ratios
reaction (2) increases the valence of metals (Me) such as Fe of ~10 appropriate for these magmas (e.g., as little as 0.5 wt %
and Eu, as illustrated by the reaction Fe and 500 ppm S, Fe3+/Fe2+ = 0.2−0.5; 750°C; experimen-
tal data of Kress and Carmichael, 1991; Wilke and Behrens,
Me2+O + 0.25 O2 = Me3+O1.5. (4)
1999), degassing of SO2 can oxidize all the Fe in the melt. In
For example, at 750°C the experimental data on tonalite contrast, at higher Fe/S ratios (~100) typical of nonmineral-
of Wilke and Behrens (1999) demonstrate that Eu3+/Eu2+ izing more mafic and higher temperature magmas, evolution
increases as a function of log fO2: of SO2-bearing vapor minimally increases magmatic fO2 by <1
log units (Fig. 5A).
log (Eu3+/Eu2+) = 0.25 log fO2) – 3.5. (5)
Porphyry ore deposits are produced by hydrothermal fluids
The evolution of SO2 gas produces magmatic oxidation of Fe that separate during late-stage crystallization from a variety of
and Eu that can be modeled using the 750°C data of Wilke S-rich magmas with a range of compositions and Fe/S (Dilles,
248 SCIENTIFIC COMMUNCIATIONS

30 2

Plagioclase with a small Eu anomaly

6 20
1

wt.% S (as sulfate) in melt + anhydrite

(-3)
5 0.7
Calculated Evolution Fe/S = 10 Fe/S = 20 10
Paths of Silicic Melt
log f O2 plotted as ∆NNO

0.5
Containing Sulfur as Anhydrite
4 0.4
Resulting from Reaction

Fe3+/ Fe2+
SO3 + 2Fe2+O = SO2 + 2Fe3+O1.5 5

Eu3+/ Eu2+
3 0.3
0.2
Fe/S = 100

2 0.2 2
elt
S in m 0.1
(-4)
1 0.1 Vapor Saturation
1
A (H2O + SO2 + Metal-Cl)
0.05
Plagioclase with a large Eu anomaly
0 0
0.6
0 10 20 30 40 50 60 70 80 90 100
Melt Percentage Crystallized

( approximate log[ Ce4+/ Ce3+])

1.0
Field of Anomalous
Mag Eu/Eu* of Zircon
ma R
e char in Mineralizing Magmas
ge/M
ix ing
10
EuN/EuN* (zircon)

low Fe/S

Eu3+/ Eu2+ Melt

No
rm intermed. Fe/S
al
0.5 Ma
gm
ati Oxidation due to SO2 gas loss
c Ev
olu high Fe/S
tion
(little 1
SO
2 gas s
epara
tion)
0.1
B Approximate Melt Percentage Crystallized
0 10 20 30 40 50 60 70 80 90 100
8000 16000
Approximate Hf content of zircon (ppm) (absolute values vary by magmatic system)

Fig. 5. A. Melt crystallization model (after Candela and Holland, 1986; Cline and Bodnar, 1991) for change in melt oxygen
fugacity (plotted on left relative to the NNO buffer, e.g., DNNO), and melt species (Fe3+/Fe2+, Eu3+/Eu2+, and (Ce4+/Ce3+),
plotted on right as the result of vapor saturation and evolution of SO2-rich gas from anhydrite-saturated magmas (Pallister
et al., 1996; Streck and Dilles, 1998; Chambefort et al., 2008) as a function of the ratio of Fe/S in melt containing anhydrite
crystals. Model assumes that a magma with 3 wt % water at 200 MPa and initial oxidation state of DNNO + 2 must crystallize
about 50% to reach saturation in a water-rich phase into which SO2 partitions with further crystallization, that the sulfur and
iron partition coefficients are DSxl/m = 0.005, DSv/m = 20, DFexl/m ≅ DFev/m ≅ 1 (where, m =silicate melt, xl = nonanhydrite
crystals, v = vapor; Candela and Piccoli, 1995; Audétat et al., 2008). “S in melt” plots wt % sulfur in melt + anhydrite. Iron
and Eu oxidation state from experimental data on tonalite at 750°C (Wilke and Behrens, 1999). B. Zircon EuN/EuN* vs. Hf
behavior based on degassing model in (A) and observed zircon compositions. Arrows indicate trends of magmatic crystalliza-
tion/cooling and potential high-temperature recharge.

1987; Seedorff et al., 2005; Lee, 2008; Longo et al., 2010). 1999), and that in strongly oxidized conditions (Eu3+/Eu2+ >
In the Burnham (1979) model, most magmatic-hydrothermal 200) plagioclase will not preferentially incorporate Eu. Mod-
fluid is released during isobaric cooling and crystallization eling suggests that at low Fe/S of 10 to 20, SO2 gas evolution
(second boiling), corresponding to ~5 wt % H2O for horn- can produce extreme oxidation states as much as DNNO +
blende-bearing magmas. Additional water and SO2 could be 5 and Eu3+/Eu2+ >10 that are consistent with late-magmatic
contributed from degassing of underplated mafic magma, but breakdown of ilmenite to rutile plus hematite in the miner-
such gases would not produce oxidation via equation (2) (e.g., alizing Luhr Hill pluton at Yerington (Dilles, 1987). Conse-
Pinatubo, Pallister et al., 1996). Therefore, the arguments quently, evolution of SO2-rich vapor at low Fe/S ratios during
above let us model the inferred zircon EuN/EuN* values as crystallization of a granite eutectic assemblage of quartz, pla-
a function of the Hf content proxy for melt temperature and gioclase, and alkali feldspar oxidizes Eu and will not reduce
crystallization assuming that Fe/S ratio of melt governs the the EuN/EuN* in the remaining melt or in any zircon crys-
oxygen fugacity (Fig. 5B). The experimental data on tonalite tallized. Nonetheless, zircons crystallized under oxidized
melt at 750°C indicate that the Eu3+/Eu2+ ratio of melt is 15 conditions have only modest EuN/EuN* anomalies resulting
times greater than the Fe3+/Fe2+ ratio (Wilke and Behrens, from earlier crystallization of plagioclase and are predicted to
 SCIENTIFIC COMMUNCIATIONS 249

evolve to high Hf content at constant EuN/EuN*, as illustrated Regardless of the origin, the REE geochemistry of zircon is a
in Figure 5B. The origin of the characteristic high variability potential fingerprint of ore-forming intrusions that could be
of EuN/EuN* in mineralized zircons is not known (Fig. 4). One applied in regional samples of plutons or detrital zircons as an
mechanism may be recharge and mixing of high-temperature exploration tool.
and crystal-poor magmas with the low-temperature oxidized
and degassed magmas (Fig. 5B). In any case, the relationship Acknowledgments
between EuN/EuN* and Hf in zircons may provide a record of This research was materially supported by grant EAR-
magmatic Fe/S ratios and help to identify ancient magmas that 1049792 from the National Science Foundation and gifts from
were sufficiently S rich to form ore deposits. Because zircon several minerals companies (CODELCO, Freeport McMo-
is a highly resistant mineral during hydrothermal alteration Ran, American Barrick, Newmont, and the Pebble Partner-
and in surficial environments, its composition can potentially ship). Brad Ito and the SUMAC staff kept the SHRIMP-RG
be used in regional surveys to locate S-rich magmatic bodies running at peak performance and efficiency. We are thank-
capable of forming significant ore deposits. ful for help from and discussions with Frank Mazdab, Anita
The array of natural zircon EuN/EuN* versus Hf data in Grunder, B. J. Walker, Frank Tepley, Nansen Olson, Matthew
mineralizing porphyry intrusions varies for each magmatic Coble, Cyril Chelle-Michou, and the OSU VIPER group.
system studied, and we argue that this reflects the oxidation We thank J.E. Wright, S. Wyld, and J. Aleinikoff for use of
of melts with varied Fe/S during evolution of SO2-rich vapor zircon trace element data from nonmineralized plutons.
(Fig. 4). For example, some premineral intrusions (Los Picos, Jeremy Richards, Adam Simon, David John, and an anony-
McLeod Hill, and Bear at Yerington, and Butte Granite) fol- mous referee provided thoughtful reviews of versions of this
low paths of decreasing EuN/EuN* consistent with either low manuscript.
S contents and high Fe/S ratios or possibly SO2-degassing
at elevated temperatures. Early andesite porphyry magmas Author Contributions
from El Salvador associated with minor sulfide ores also have J.H.D., A.J.R.K., and J.LW. wrote this manuscript. J.H.D.
this pattern. In contrast, the late andesite porphyries related formulated the degassing model, and J.L.W. managed all lab-
to the main Cu-Mo sulfide ores at El Salvador have higher oratory analyses. J.H.D. and J.L.W. contributed to all aspects
EuN/EuN* that decreases with increasing Hf in a trend con- of the data collection, with analyses by R.G.L. and R.M.T.
sistent with SO2-rich vapor evolution at high Fe/S ratios in (El Salvador), A.K. (Yanacocha), L.P.F. (Carlin-BM). R.G.L.,
melts (Fig. 4C). Andesite to dacite magmas from Yanacocha J.H.D., R.M.T. and J.L.W. made initial identification of REE
are anhydrite bearing (Chambefort et al., 2008), have EuN/ systematics of zircons from ore-forming intrusions.
EuN* that increases slightly with Hf, and are consistent with
an intermediate Fe/S ratio of the melts. The zircons in the REFERENCES
most silicic porphyry melts from El Abra and Yerington have Annen, C., Blundy, J.D., and Sparks, R.S.J., 2006, The genesis of intermedi-
EuN/EuN* that increases with Hf consistent with SO2-rich ate and silicic magmas in deep crustal hot zones: Journal of Petrology, v. 47,
vapor evolution at low Fe/S. For many mineralizing intrusions p. 505−539.
Audétat, A., Pettke, T., and Dolejš, D., 2004, Magmatic anhydrite and calcite
the large range of EuN/EuN* at a particular Hf content is con- in the ore-forming quartz-monzodiorite magma at Santa Rita, New Mexico,
sistent with complex multicyclic magma recharge and mixing USA: Genetic constraints on porphyry-Cu mineralization: Lithos, v. 72, p.
followed by cooling, crystallization, and degassing in one or 147−161.
more shallow magma chambers (Longo et al., 2010; Chambe- Audétat, A., Pettke, T., Heinrich, C.A., and Bodnar, R.J., 2008, The composi-
tion of magmatic-hydrothermal fluids in barren and mineralized intrusions:
fort et al., 2013; Fig. 5B). Economic Geology, v. 103, p. 877−908.
Ballard, J.R., Palin, J.M., and Campbell, I.H., 2002, Relative oxidation states
Conclusions of magmas inferred from Ce(IV)/Ce(III) in zircon: Application to porphyry
As recognized by Ballard et al. (2002) and other workers, zir- copper deposits of northern Chile: Contributions to Mineralogy and Petrol-
con with EuN/EuN* >0.4 is characteristic of many ore-forming ogy, v. 144, p. 347–364.
Baker, L.L., and Rutherford, M.J., 1996, Crystallisation of anhydrite-bearing
arc magmas. Plots of EuN/EuN* versus Hf in zircons in most magmas: Geological Society of America Special Paper 315, p. 243–250.
cases illustrate distinctive arrays for ore-forming arc intrusions Bell, A.S., and Simon, A., 2011, Experimental evidence for the alteration of
compared to nonore-forming intrusions both within the same the Fe3+/SFe of silicate melt caused by the degassing of chlorine-bearing
magmatic suite as the ore-forming intrusions and regionally aqueous volatiles: Geology, v. 39, p. 499−502.
Brennicka, G., 2006, Origin and metal content of magmatic sulfides in Cu-Au
within suites of the same age. Mineralizing intrusions contain mineralizing silicic magmas: Yerington, Nevada, and Yanacocha, Peru: M.S.
higher and strikingly variable EuN/EuN* compared to earlier thesis, Corvallis, OR, Oregon State University, 49 p. http://hdl.handle.net/
nonmineralized intrusions of the same magmatic suite and 1957/2076.
regionally. We propose that the zircon REE geochemistry Burgisser, A., and Scaillet, B., 2007, Redox evolution of a degassing magma
trends of ore-related intrusions reflect both high water con- rising to the surface: Nature, v. 445, p. 194–197, doi: 10.1038/nature05509.
Burnham, A.D., and Berry, A.J., 2012, An experimental study of trace ele-
tent and consequent suppression of early plagioclase crystal- ment partitioning between zircon and melt as a function of oxygen fugacity:
lization, as well as late magmatic oxidation attending loss of Geochimica et Cosmochimica Acta, v. 95, p. 196–212.
SO2-rich magmatic-hydrothermal ore fluids during late-stage Burnham, C.W., 1979, Magmas and hydrothermal fluids, in Barnes, H.L., ed.,
crystallization of granites. The Fe/S ratio of melts influences Geochemistry of hydrothermal ore deposits: New York, Wiley, p. 71–136.
late oxidation because increased degree of oxidation reflected Burnham, C.W., and Ohmoto, H., 1980, Late stage processes of felsic mag-
matism: Society of Mining Geologists Japan Special Issue 8, p. 1−11.
in small EuN/EuN* anomalies of zircon is associated with low Candela, P.A., 1986, The evolution of aqueous vapor from silicate melts:
Fe/S of melt. This hypothesis reinforces the key role of water Effect on oxygen fugacity: Geochimica et Cosmochimica Acta, v. 50, p.
and sulfur in mineralizing magmas, and merits future testing. 1205–1211, doi: 10.1016/0016-7037(86)90403-5.
250 SCIENTIFIC COMMUNCIATIONS

——1989, Felsic magmas, volatiles and metallogenesis: Reviews in Economic Main stage base mineral deposits, Butte district, Montana: Chemical Geol-
Geology, v. 4, p. 223−233. ogy, v. 215, p. 61−93.
Candela, P.A., and Holland, H.D., 1986, A mass transfer model for copper Fohey-Breting, N.K., Barth, A.P., Wooden, J.L., Mazdab, F.K., Carter, C.A.,
and molybdenum in magmatic hydrothermal systems: The origin of por- and Schermer, E.R., 2010, Relationship of voluminous ignimbrites to con-
phyry-type ore deposits: Economic Geology, v. 81, p. 1−19. tinental arc plutons: Petrology of Jurassic ignimbrites and contemporane-
Candela, P.A., and Piccoli, P., 1995, Model ore-metal partitioning from melts ous plutons in southern California: Journal of Volcanology and Geothermal
into vapor and vapor/brine mixtures: Mineralogical Association of Canada, Research, v. 189, p. 1−11.
Short Course, v. 23, p.101−127. Fu, B., Mernagh, T.P., Kita, N.T., Kemp, A.I.S., and Valley, J.W., 2009, Distin-
Carn, S. A., and Prata, F.J., 2010, Satellite-based constraints on explosive SO2 guishing magmatic zircon from hydrothermal zircon: A case study from the
release from Soufrière Hills Volcano, Montserrat: Geophysical Research Gidginbung high-sulphidation Au-Ag-(Cu) deposit, SE Australia: Chemical
Letters, v. 37, 5 p. L00E22, doi:10.1029/2010GL044971. Geology, v. 259, p. 131–142.
Carroll, M.R., and Rutherford, M.J., 1987, The stability of igneous anhydrite: Gustafson, L.B., 1979, Porphyry copper deposits and calc-alkaline volcanism,
Experimental results and implications for sulfur behavior in the 1982 El in McElhinny, M.W., ed., The Earth: Its origin, structure, and evolution,
Chichón trachyandesite and other evolved magmas: Journal of Petrology, Academic Press, London, p. 427−468.
v. 28, p. 781–801. Gustafson, L.B., and Hunt, L. B., 1975, The porphyry copper deposit at El
Chambefort, I., Dilles, J.H., and Kent, A.J.R., 2008, Anhydrite-bearing Salvador, Chile: Economic Geology, v. 70, p. 857−912.
andesite and dacite as a source for sulfur in magmatic-hydrothermal min- Gustafson, L.B., Orquera, W., McWilliams, M., Castro, M., Olivares, O.,
eral deposits: Geology, v. 36, p. 719−722. Rojas, G., Maluenda, J., and Mendez, M., 2001, Multiple centers of miner-
Chambefort, I., Dilles, J.H., and Longo, A.A., 2013, Amphibole geochemistry alization in the Indio Muerto district, El Salvador, Chile: Economic Geol-
of the Yanacocha Volcanics, Peru: Evidence for diverse sources of magmatic ogy, v. 96, p. 325−350.
volatiles related to gold ores: Journal of Petrology, v. 54, p. 1017−1046. Halter, W.E., Heinrich, C.A., and Pettke, T., 2004, Laser-ablation ICP-MS
Chiaradia, M., Merino, D., and Spikings, R., 2009, Rapid transition to analysis of silicate and sulfide melt inclusions in an andesitic complex II:
long-lived deep crustal magmatic maturation and the formation of giant Evidence for magma mixing and magma chamber evolution: Contributions
porphyry-related mineralization (Yanacocha, Peru): Earth and Planetary to Mineralogy and Petrology, v. 147, p. 397–412.
Science Letters, v. 288, p. 505–515. Hildreth, W., and Moorbath, S., 1988, Crustal contributions to arc magma-
Claiborne, L.L., Miller, C.F., and Wooden, J.L., 2010, Trace element compo- tism in the Andes of central Chile: Contributions to Mineralogy and Petrol-
sition of igneous zircon: A thermal and compositional record of the accu- ogy, v. 98, p.455−489.
mulation and evolution of a large silicic batholith, Spirit Mountain, Nevada: Hoskin, P.W.O., and Schaltegger, U., 2003, The composition of zircon and
Contributions to Mineralogy and Petrology, v. 160, p. 511−531. igneous and metamorphic petrogenesis: Reviews in Mineralogy and Geo-
Cline, J., 1995, Genesis of porphyry copper deposits: The behavior of water, chemistry, v. 53, p. 27−62.
chloride, and copper in crystallizing melts: Arizona Geology Digest, v. 20, Houston, R.A., and Dilles, J.H., 2013, Structural geologic evolution of the
p. 69−82. Butte district, Montana: Economic Geology, v. 108, p. 1397−1424.
Cline, J.S., and Bodnar, R.J., 1991, Can economic porphyry copper mineral- Jugo, P.J., 2009, Sulfur content at sulfide saturation in oxidized magmas:
ization be generated by a typical calc-alkaline melt?: Journal of Geophyscial Geology, v. 37, p. 415−418.
Research, v. 96, p. 8113−8126. Jugo, P., Luth, R., and Richards, J., 2005, An experimental study of the sulfur
Claiborne, L.L., Miller, C.F., and Wooden, J.L., 2010, Trace element compo- content in basaltic melts saturated with immiscible sulfi de or sulfate liquids
sition of igneous zircon: A thermal and compositional record of the accu- at 1300°C and 1.0 GPa: Journal of Petrology, v. 46, p. 783–798.
mulation and evolution of a large silicic batholith, Spirit Mountain, Nevada: Kay, S.M., and Mpodozis, C., 2001, Central Andean ore deposits linked to
Contributions to Mineralogy and Petrology, v. 160, p. 511−531. evolving shallow subduction systems and thickening crust: GSA Today, v.
Czehura, S., 2006, Butte: A world class ore deposit: Mining Engineering, v. 11, p. 4−9.
58, p. 14–19. Kelley, K.A., and Cottrell, E., 2009, Water and the oxidation state of subduc-
Deen, J.A., Rye, R.O., Munoz, J.L., and Drexler, J.W., 1994, The magmatic tion zone magmas: Science, v. 325, p. 605–607.
hydrothermal system at Julcani, Peru: Evidence from fluid inclusions and Keppler, H., 1999, Experimental evidence for the source of excess sulfur in
hydrogen and oxygen isotopes: Economic Geology, v. 89, p. 1924−1938. explosive volcanic eruptions: Science, v. 284, p. 1652−1654.
Defant, M.J., and Drummond, M.S., 1990, Derivation of some modern arc ——2010, The distribution of sulfur between haplogranitic melts and aque-
magmas by melting of young subducted lithosphere: Nature, v. 347, p. ous fluids: Geochimica et Cosmochimica Acta, v. 74, p. 645–660.
662–66. Kress,V., and Carmichael, I.S.E., 1991, The compressibility of silicate liq-
Dilles, J., 1987, Petrology of the Yerington batholith, Nevada: Evidence for uids containing Fe2O3 and the effect of composition, temperature, oxygen
evolution of porphyry copper ore fluids: Economic Geology, v. 82, p. fugacity, and pressure on their redox states: Contributions to Mineralogy
1750–1789. and Petrology, v. 108, p. 82–92.
Dilles, J.H., and Einaudi, M.T., 1992, Wall-rock alteration and hydrother- Lee, C.-T.A., Luffi, P., Chin, E. J., Bouchet, R., Dasgupta, R., Morton, D. M.,
mal flow paths about the Ann-Mason porphyry copper deposit, Nevada—a LeRoux, V., Yin, Q.-Z., and Jin, D., 2012, Copper systematics in arc magmas
6-km vertical reconstruction: Economic Geology, v. 87, p. 1963−2001. and implications for crust-mantle differentiation: Science, v. 336, p. 64–68.
Einaudi, M.T., Hedenquist, J.W., and Inan, E.E., 2003, Sulfidation state of Lee, R.G., 2008, Genesis of the El Salvador porphyry copper deposit, Chile
fluids in active and extinct hydrothermal systems; transitions from porphyry and distribution of epithermal alteration at Lassen Peak, California: Ph.D.
to epithermal environments: Society of Economic Geologists Special Pub- thesis, Corvallis, OR, Oregon State University, 322 p.
lication 29, p. 285−313. Liang, H.Y., Campbell, I.H., Allen, C., Sun, W.D., Liu, C.Q., Yu, H.X., Xie,
Farmer, L.P., 2013, Trace element characteristics of zircon: A means of Y.W., and Zhang, Y.Q., 2006, Zircon Ce4+/Ce3+ ratios and ages for Yulong
assessing mineralization potential of intrusions in northern Nevada: M.S. ore-bearing porphyries in eastern Tibet: Mineralium Deposita, v. 41, p.
thesis, Corvallis, OR, Oregon State University, 92 p. http://hdl.handle. 152–159.
net/1957/36276 Lipske, J.L., and Dilles, J.H., 2000, Advanced argillic and sericitic alteration
Ferriss, E.D.A., Essene, E.J., and Becker, U., 2008, Computational study of in the subvolcanic environment of the Yerington porphyry copper district,
the effect of pressure on the Ti-in-zircon geothermometer: European Jour- Buckskin Range, Nevada: Society of Economic Geologists Guidebook 32,
nal of Mineralogy, v. 20, p. 745−755. p. 91−100.
Ferry, J.M., and Watson, E.B., 2007, New thermodynamic models and revised Longo, A.A., Dilles, J.H., Grunder, A.L., and Duncan, R., 2010, Evolution of
calibrations for the Ti-in-zircon and Zr-in-rutile thermometers: Contribu- calc-alkaline volcanism and associated hydrothermal gold deposits at Yana-
tions to Mineralogy and Petrology, v. 154, p. 429–437. cocha, Perú: Economic Geology, v. 105, p. 1191−1241.
Fiege, A., Behrens, H., Holtz, F., and Adams, F., 2014, Kinetic vs. thermody- López, T., Ushako, S., Izbekov, P., Tassi, F., Cahill, C., Neill, O., and Wer-
namic control of degassing of H2O-S ± Cl-bearing andesitic melts: Geochi- ner, C., 2013, Constraints on magma processes, subsurface conditions, and
mica Cosmochima Acta, v. 125, p. 241−264. total volatile flux at Bezymianny Volcano in 2007–2010 from direct and
Field, C.W., Zhang, L., Dilles, J.H., Rye, R.O., and Reed, M.H., 2005, Sulfur remote volcanic gas measurements: Journal of Volcanology and Geothermal
and oxygen isotopes of pre-Main stage porphyry Cu-Mo and late shallow Research, v. 263, p. 92–107.
 SCIENTIFIC COMMUNCIATIONS 251

Loucks, R.R., 2014, Distinctive composition of copper-ore-forming arc mag- Scaillet, B., Clemente, B., Evans, B.W., and Pichavant, M., 1998, Redox con-
mas: Australian Journal of Earth Sciences, v. 61, p. 5−16. trol of sulfur degassing in silicic magmas: Journal of Geophysical Research,
Luhr, J. F., 1990, Experimental phase relations of water- and sulfur-saturated v. 103 (B1), p. 23,937−23,949.
arc magmas and the 1982 eruptions of El Chichón volcano: Journal of Seedorff, E., Dilles, J.H., Proffett, J.M., Jr., Einaudi, M.T., Zurcher, L., Wil-
Petrology, v. 31, p. 1071–1114. liam J. A. Stavast, W. J.A., Johnson, D.A., and Barton, M.D., 2005, Por-
Lund, K., Aleinikoff, J.N., Kunk, M.J., Unruh, D.M., Zeihen, G.D., Hodges, phyry-related deposits: Characteristics and origin of hypogene features:
W.C., Du Bray, E.A., and O’Neill, J.M., 2002, SHRIMP U-Pb and 40Ar/39Ar Economic Geology 100TH Anniversary Volume, p. 251−298.
age contraints for relating plutonism and mineralization in the Boulder Sillitoe, R.H., 2010, Porphyry copper systems: Economic Geology, v. 105,
batholith region, Montana: Economic Geology, v. 97, p. 241−267. p. 3−41.
Mazdab, F.K., and Wooden, J.L., 2006, Trace element analysis in zircon by Simmons, S.F., White, N.C., and John, D.A., 2005, Geological characteris-
ion microprobe (SHRIMPRG): Technique and applications: Geochimica et tics of epithermal precious and base metal deposits: Economic Geology
Cosmochimica Acta Supplement, v. 70, p. 405. 100TH Anniversary Volume, p. 485−522.
Mungall, J.E., 2002, Roasting the mantle: Slab melting and the genesis of Streck, M.J., and Dilles, J.H., 1998, Sulfur content of oxidized arc magmas
major Au and Au-rich Cu deposits: Geology, v. 30, p. 915–918. as recorded in apatite from a porphyry copper batholith: Geology, v. 26, p.
Muñoz, M., Charrier, R., Fanning, C.M., Maksaev, V., and Deckart, K., 2012, 523−526.
Zircon trace element and O-Hf isotope analyses of mineralized intrusions Sun, W.D., Ling, M.X., Chung, S.L., Ding, X., Yang, X.Y., Liang, H.Y., Fan,
from El Teniente ore deposit, Chilean Andes: Constraints on the source W.M., Goldfarb, R., and Yin, Q.Z., 2012, Geochemical constraints on ada-
and magmatic evolution of porphyry Cu-Mo related magmas: Journal of kites of different origins and copper mineralization: Journal of Geology, v.
Petrology, v. 53, p. 1091–1122. 120, p. 105–120.
Muntean, J.L., Cline, J.S., Simon, A.C., and Longo, A.A., 2011, Magmatic- Tosdal, R.M., Dilles, J.H., and Cooke, D.R., 2009, From source to sinks in
hydrothermal origin of Nevada’s Carlin-type gold deposits: Nature Geosci- auriferous magmatic-hydrothermal porphyry and epithermal deposits: Ele-
ence, v. 4, p. 122−127. ments, v. 5, p. 289−29.
Pallister, J.S., Hoblitt, R.P., and Reyes, A.G., 1992, A basalt trigger for the Trail, D., Watson, E.B., and Tailby, N.D., 2011, The oxidation state of Hadean
1991 eruptions of Pinatubo volcano?: Nature, v. 356, p. 426−428. magmas and implications for early Earth’s atmosphere: Nature, v. 480, p.
Pallister, J.S., Hoblitt, R.P., Meeker, G.P., Knight, R.J., and Siems, R.J., 1996, 79−82.
Magma mixing at Mount Pinatubo: Petrographic and chemical evidence Van Hoose, A.E., Streck, M.J., Pallister, J.S., and Wälle, M., 2013, Sulfur evo-
from the 1991 deposits, in Newhall, C.G., and Punongbayan, R.S., eds., lution of the 1991 Pinatubo magmas based on apatite: Journal of Volcanol-
Fire and mud: Eruptions and lahars of Mount Pinatubo, Philippines: Seat- ogy and Geothermal Research, v. 257, p. 72–89.
tle, WA, University of Washington Press, p. 687−731. Wainwright, A.J., Tosdal, R.M. Wooden, J.L., Mazdab, F.K., and Friedman,
Parat, F., and Holtz, F., 2005, Sulfur partition coefficient between apatite R.M, 2011, U-Pb (zircon) and geochemical constraints on the age, origin,
and rhyolite: The role of bulk S content: Contributions to Mineralogy and and evolution of Paleozoic arc magmas in the Oyu Tolgoi porphyry Cu-Au
Petrology, v. 150, p. 643−651. district, southern Mongolia: Gondwana Research, v. 19, p. 764–787.
Pokrovski, G. S., Borisova, A.Y., and Harrichoury, J-C., 2008, The effect of Wallace, P.J., and Edmonds, M., 2011, The sulfur budget in magmas; evi-
sulfur on vapor-liquid fractionation of metals in hydrothermal systems: dence from melt inclusions, submarine glasses, and volcanic gas emissions:
Earth and Planetary Science Letters, v. 266, p. 345–362. Reviews in Mineralogy and Geochemistry, v. 73, p. 215−246.
Reed, M.J., Candela, P.A., and Piccoli, P.M., 2000, The distribution of rare Wang, F., Liu, S-A; Li, S., and He, Y., 2013, Contrasting zircon Hf-O isotopes
earth elements between monzogranite melt and the aqueous volatile phase and trace elements between ore-bearing and ore-barren adakitic rocks in
in experimental investigations at 800°C and 200 MPa: Contributions to central-eastern China: Implications for genetic relation to Cu-Au mineral-
Mineralogy and Petrology, v. 140, p. 251–262. ization: Lithos, v. 156−159, p. 97−111.
Richards, J.P., 2003, Tectono-magmatic precursors for porphyry Cu-(Mo-Au) Watson, E.B., and Harrison, T.M., 2005, Zircon thermometer reveals mini-
deposit formation: Economic Geology, v. 98, p. 1515–1533, doi:10.2113/ mum melting conditions on earliest earth: Science, v. 308, p. 841−844.
gsecongeo.98.8.1515. Whitney, J.A., 1988, Composition and activity of sulfurous species in
——2011a, High Sr/Y arc magmas and porphyry Cu ± Mo ± Au deposits: Just quenched magmatic gases associated with pyrrhotite-bearing silicic sys-
add water: Economic Geology, v. 106, p. 1075−1081. tems: Economic Geology, v. 83, p. 86−92.
——2011b, Magmatic to hydrothermal metal fluxes in convergent and col- Wilke, M., and Behrens, H., 1999, The dependence of the partitioning of iron
lided margins: Ore Geology Reviews, v. 40, p. 1–26. and europium between plagioclase and hydrous tonalitic melt on oxygen
Richards, J.P., Spell, T., Rameh, E., Razique, A., and Fletcher, T., 2012, High fugacity: Contributions to Mineralogy and Petrology, v. 137, p. 102−114.
Sr/Y magmas reflect arc maturity, high magmatic water content, and por- Wooden, J.L., Mazdab, F.K., Mueller, P.A., Wiegand, B., Kita, N.T., and Val-
phyry Cu ± Mo ± Au potential: Examples from the Tethyan arcs of central ley, John W., 2008, Geochemical and isotopic evidence for the origin of the
and eastern Iran and western Pakistan: Economic Geology, v. 107, p. Boulder batholith, Montana: Geochimica et Cosmochimica Acta, v. 72, p.
295–332. A1034.
Rohrlach, B.D., and Loucks R.R., 2005, Multi-million-year cyclic ramp-up of Zajacz, Z., Candela, P.A., Piccoli, P.M., and Sanchez-Valle, C., 2012 The
volatiles in a lower crustal magma reservoir trapped below the Tampakan partitioning of sulfur and chlorine between andesite melts and magmatic
Cu-Au deposit by Mio-Pliocene crustal compression in the southern Philip- volatiles and the exchange coefficients of major cations: Geochimica et Cos-
pines, in Porter, T.M., ed., Super porphyry copper and gold deposits: Porter mochimica Acta, v. 89, p.81–101.
Geoconsultancy Publishing, v. 2, p. 369−407.