2014 Society of Economic Geologists, Inc.

Economic Geology, v. 109, pp. 17051733

High-Sulfidation Epithermal Pyrite-Hosted Au (Ag-Cu) Ore Formation

by Condensed Magmatic Vapors on Sangihe Island, Indonesia*
Julia King,1, A.E. Williams-Jones,1 Vincent van Hinsberg,1 and Glyn Williams-Jones2
1Department of Earth and Planetary Sciences, McGill University, 3450 University Street, Montral, Qubec, Canada H3A 0E8
2Department of Earth Sciences, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, Canada V5A 1S6

Abstract
Although gold in high-sulfidation epithermal deposits generally occurs as the native metal or electrum, in
some deposits, a significant proportion of the gold is hosted in pyrite. Here we use a combination of petrog-
raphy, whole-rock geochemistry, pyrite chemistry, crystallography, and phase stability relationships to deter-
mine how gold was transported and incorporated into pyrite in two relatively young high-sulfidation epithermal
deposits, where the gold occurs almost exclusively in solid solution or as nanoparticles in pyrite.
The genetically related Bawone and Binebase Au (Cu-Ag) deposits, located 1 km apart on the volcanic island
of Sangihe, northeastern Indonesia, are hosted by andesitic volcaniclastic rocks that were altered to a proximal
advanced argillic association of quartz + pyrite (py I) + pyrophyllite + natroalunite + alunite + dickite + kaolin-
ite and a more distal intermediate argillic association of quartz + pyrite (py I) + kaolinite + dickite + illite. The
economic mineralization takes the form of multiple generations of auriferous pyrite, the first of which, pyrite
I (py I), developed during advanced argillic alteration. Mass balance calculations show that all elements were
mobile with the exception of Nb, Ti, some rare earth elements, and possibly Al.
The highest gold concentration is in pyrite II (py II), which occurs in veins that cut pyrite I. This drusy vari-
ety of pyrite is characterized by complex growth and sector zoning, and contains as much as 6.0 wt % Cu. The
elevated Cu concentrations correlate positively with Au and As concentrations, whereas the Ag concentration
correlates strongly with Au but not Cu. Later barite-enargite mineralization exploited py II veins and vugs, and
significant concentrations of Ag and Au are hosted by enargite, although the Au concentration in enargite is
lower than in py II or py I.
A model is presented in which the fluid responsible for advanced argillic and intermediate argillic alteration
and associated stage 1 gold mineralization was a condensed magmatic vapor derived from an oxidized magma.
The gold and other metals were transported as hydrated species that ascended through the volcanic pile via
fractures and zones of enhanced permeability to a depth between 900 and 1300 m, where the vapor condensed
at a temperature between 250 and 340C to form an acidic liquid with a pH of ~2.5; fO2 ranged up to four log
units above the hematite-magnetite buffer. Interaction of this liquid with the host andesites caused advanced
argillic and intermediate argillic alteration, including sulfidation of mafic minerals to form py I. During crystal-
lization of py I, Au, Cu and Ag were adsorbed onto the surface of the pyrite and deposited as nanoparticles, or
were incorporated in the pyrite structure. Adsorption of Au, Cu, and Ag from the condensed vapor reached a
peak during the crystallization of vein-hosted py II, and the uptake of Ag and minor Au continued during later
crystallization of enargite. From the distribution of metals among growth and sector zones in py II, incorpora-
tion of gold and other metals appears to have been maximized when physicochemical conditions were relatively
stable. This is in contrast to the requirement for native gold precipitation, namely that physicochemical gradi-
ents be steep to ensure supersaturation of gold in the ore fluid.

Introduction metals (Stoffregen, 1987; White and Hedenquist, 1990; Arri-

Although considerable progress has been made in under-
standing the formation of high-sulfidation epithermal pre-
cious metal deposits, opinion is still divided over the nature
of the ore fluid. There is general agreement that the char-
acteristic residual (vuggy) silica and advanced argillic altera-
tion are the result of interaction of rocks with the condensate
of highly acidic and oxidizing vapors (e.g., HCl, and H2S and
H2SO4 produced by reaction of H2O with SO2) from a proxi-
mal magma source (e.g., Hemley and Jones, 1964; Stoffregen
1987; Rye, 1993; Arribas 1995). However, until recently, most
researchers, noting that the ore minerals commonly fill vugs
and therefore postdate the alteration, have concluded that the
fluid responsible for the alteration did not transport the ore
Corresponding
author: e-mail, julia.j.king@gmail.com
*A digital supplement to this paper, containing four Appendices with raw
data, is available at http://economicgeology.org/ and at http://econgeol.geo-
scienceworld.org/.
0361-0128/14/4246/1705-29 1705
bas, 1995; Hedenquist et al., 1998). Instead, these researchers
have attributed the mineralization to collapse of the vapor-
dominated system and transport of the metals by a magmatic-
hydrothermal liquid with a large meteoric water component.
The discovery that gold mineralization in the Pascua deposit,
Chile, was contemporaneous with advanced argillic alteration
(Chouinard et al., 2005a), indicates that for some high-sulfida-
tion epithermal deposits (cf. Voudouris, 2010) the two-stage
hydrothermal model does not apply. This, and a combina-
tion of fluid inclusion (Heinrich et al., 1999; Landtwing et
al., 2010) and experimental evidence (Archibald et al., 2001,
2002; Williams-Jones et al., 2002; Zezin et al., 2011b; Migdisov
and Williams-Jones, 2013; Hurtig and Williams-Jones, 2014)
showing that gold, silver, and copper may be considerably
more soluble in aqueous vapors than previously suspected,
supports a model for these deposits in which hydrothermal
alteration and economic mineralization were both products
of a magmatic hydrothermal vapor (cf. Williams-Jones and
Submitted: March 31, 2013
Accepted: December 9, 2013
1706 KING ET AL.
Heinrich, 2005; Mavrogenes et al., 2010; Berger and Henley,
2011; Henley and Berger, 2011; Scher et al., 2013).
Most studies of high-sulfidation epithermal systems have
focused on deposits in which the gold occurs as a discrete
mineral or minerals (e.g., native gold, electrum, and/or a tellu-
ride phase, such as calaverite; Kesler et al., 1981; Stoffregen,
1987; Moritz et al., 2004; Deditius et al., 2008). However, in
some high-sulfidation deposits, notably the Pascua deposit in
Chile, a high proportion of the gold is hosted in sulfide miner-
als, particularly pyrite (e.g., Chouinard et al., 2005b; Deditius
et al., 2009). Here, we report results of a study of two high-
sulfidation epithermal deposits, Bawone (9.5 Mt of ore with
an average grade of 1.32 g/t Au and 3.97 g/t Ag) and Bine-
base (17.9 Mt of ore with an average grade of 0.76 g/t Au and
18.7g/t Ag), located on Sangihe Island, Indonesia, in which
virtually all the hypogene gold is hosted in pyrite. Most signifi-
cantly, greater than 50% of this gold is in pyrite that forms part
of the early advanced argillic alteration mineral association;
the rest is contained mainly in pyrite veins that cut the altered
rocks. Thus, gold mineralization was both contemporaneous
with and postdated alteration. Based on this and other obser-
vations, we develop a model designed to explain the genesis of
the Bawone and Binebase deposits involving transport of the
metals in a highly acidic vapor and sorption of the gold onto
the surfaces of growing pyrite crystals. Given the early tim-
ing of the Au-Ag mineralization and the observation that it is
hosted almost exclusively by pyrite, we also propose that these
deposits are representatives of a subclass of high-sulfidation
epithermal precious metal deposits, in which the bulk of the
metal is hosted in pyrite and the ore fluid was a condensed
magmatic vapor.
Regional Geologic Setting
The Bawone and Binebase Au (Ag-Cu) deposits are located
in the southern part of Sangihe Island, the largest of the
islands in the 500 km long Sangihe arc, which runs north-
south from southern Mindanao, Philippines, to the north arm
of Sulawesi, and separates the Celebes and Molucca Seas
(Figs. 1, 2; Hall, 2000). The Sangihe arc formed as a result of
the westerly subduction of the Molucca Sea plate under the
Eurasian plate, whereas the facing Halmahera arc was pro-
duced by the easterly subduction of the Molucca Sea plate
under the Philippine Sea plate (Morrice et al., 1983; Morrice
and Gill, 1986; Garwin, 1990; Hall, 1996, 2000). Twenty-five
Quaternary stratovolcanoes are located along the length of the
arc; eight of these are active, including Awu at the northern
tip of Sangihe Island (Morrice and Gill, 1986; Fig. 1). Previ-
ous studies of volcanism along the Sangihe Arc have shown
that the arc is composed mainly of two-pyroxene and horn-
blende andesites. Calc-alkaline suites dominate and vary from
low to high K, depending on their distance from the volcanic
front (Morrice and Gill, 1986).
Sangihe Island has a lobate form defined by volcanic cen-
ters, the active stratovolcano, Awu, at the northern end and
the dormant/extinct stratovolcanoes, Tahuna and Kakiraeng,
and strongly weathered centers of Taware and Malisang to the
south (Fig. 1). The oldest part of the island is in the southeast
and the youngest in the northwest; new volcanoes are forming
off the northwestern tip and western shores of Sangihe Island
(Beaulieu, 2010). Given the heavy rainfall (greater than 3 m
per year), dense vegetation and inferred rapid uplift due to
the compressive tectonic regime, even the oldest volcanoes, in
the southern part of the island where the Bawone and Bine-
base deposits are located, may be only tens of thousands of
years in age.
The southern part of Sangihe Island is dominated by clino-
pyroxene andesite flows, breccias, lahars, and tuffs of the
Tamako Group, except in the southern and eastern parts
where rocks of the Taware Group, Malisang Group, Bine-
base Group, and Pinterang Formation are exposed (Fig. 2).
The Binebase deposit is 1 km to the north of Bawone, and
both deposits are hosted by rocks of the Binebase Group. At
Bawone, the Binebase Group is overlain unconformably by
the Pinterang Formation, which covers the deposit, whereas
at Binebase, the deposit crops out and is strongly oxidized to
a maximum depth of 70 m (Fig. 3). The Binebase Group has a
northeasterly strike, moderate to steep southeasterly dip, and
comprises andesitic ash and crystal tuffs, hornblende-pyrox-
ene andesite flows, biotite-hornblende-magnetite diorite and
minor dacite to rhyolite flows, which are interpreted to be
volcanic and subvolcanic facies of the extinct and eroded
Taware volcano (Garwin, 1990). To the east of the Binebase
deposit, Binebase Group rocks are overlain by rocks of the
Tamako Group (consisting of hornblende andesite flows, sills,
and dikes), the proximal facies of the dormant and/or extinct
Kakiraeng volcano, and to the south by the Pinterang For-
mation. East of the Bawone deposit, the Pinterang Forma-
tion overlies the Tamako Group rocks unconformably. This
unit is thickest (<100 m) in topographic lows, and consists of
reworked cross-bedded volcanic silts and sands, carbonates
and organic-rich sediments, which record a marine incursion
in the southeast part of Sangihe Island (Garwin, 1990). The
Pinterang Formation also contains slightly rounded, unoxi-
dized aggregates of pyrite fragments, similar to pyrite II (py
II) from the Bawone and Binebase hypogene ores. The Pin-
terang Formation and Tamako Group overlie the Malisang
Group, which in turn unconformably overlies the Binebase
Group in the southeast part of the island. Stegodon fossils
found in conglomeritic channels in the Pinterang Forma-
tion indicate that the unit is between 2 million years and
60,000years in age (de Vos et al., 2007). The Malisang Group
consists of hornblende andesite flows, sills, dikes, and diorite
intrusions that form local highs and is related to the Malisang
volcanic center (Garwin, 1990).
Local Geologic Setting
Recent oxidation has enriched both Au and Ag, and con-
sequently, much of the potentially economic mineralization
of the Binebase deposit and a part of the Bawone deposit are
supergene in origin. However, both deposits have appreciable
reserves of hypogene sulfide mineralization. The inferred
resources, using a 0.25 g/t Au cutoff, are summarized in Table
1.
The Bawone and Binebase deposits are hosted by andesites
and dacites of the Binebase Group. The Binebase deposit is
exposed at the erosional surface and has been oxidized up to
a depth of 70m (Fig. 3B). The host andesite is a crystal-rich
tuff containing 10 to 15 vol % of plagioclase laths and round
quartz crystals (12 mm in length and diameter, respec-
tively) and occasional lapilli, and lithic fragments (0.53 cm
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1707

Sangihe I.

340'0"N

Awu (1327m)

Tahuna (697m)

335'0"N
330'0"N

Kakiraeng (1006m)
Binebase (65m)
Bawone (86m)

Malisang (295m)

325'0"N
Taware (306m)

0 10
320'0"N

Km

12530'0"E 12540'0"E
Fig. 1. Inset: Location of Sangihe Island. Digitial elevation model of Sangihe Island showing the locations and elevations
of the active volcano, Awu, dormant and/or extinct volcanic centers (Tahuna, Kakiraeng, Malisang) and the Bawone and Bine-
base deposits. Bathymetry contours are at 50-m intervals.

in diameter). Rare breccia zones containing a wide variety of
rock fragments are likely epiclastic deposits.
At Bawone, a thick andesite unit overlies flow-banded pla-
gioclase-quartz-phyric dacite (Figs. 3A, 4A). In least-altered
Table 1. Inferred Resources for the Bawone and
Binebase Deposits (Stone, 2010)
Deposit Tonnes Au (g/t) Ag (g/t) Au (oz) Ag (oz)
Bawone Oxide 3,475,000 1.66 9.16 185,464 205,933
Sulfide 5,999,000 1.12 0.97 216,020 187,089
Binebase
Oxide 7,851,000 1.10 25.13 277,661 6,343,299
Sulfide 10,002,000 0.49 13.60 157, 573 4,373,443
samples, the andesite contains 5 to 30 vol % white plagio-
clase laths (2 mm long) and rare, rounded quartz phenocrysts
(2mm in diameter) in an aphanitic matrix. Conformable lay-
ers of welded crystal tuff are locally present in the andesite.
These contain 10 vol % of megascopic crystals of <3-mm-long
plagioclase laths and rare, rounded <3-mm-diameter quartz
grains. This unit is intruded by hypabyssal andesitic por-
phyry (Fig. 3A). The hypabyssal porphyry contains 20 vol%,
5 mm long, subhedral to euhedral hornblende, magnetite,
biotite and plagioclase phenocrysts, and rare quartz eyes in
an aphanitic matrix. An intrusive breccia with subangular
to subrounded, 1- to 10-cm-diameter fragments of the host
andesite and dacite generally defines the margins of the hyp-
abyssal porphyry (Fig. 4B). At Bawone, the locally overlying
1708 KING ET AL.

12535'0"E 12540'0"E

Extinct Volcano
Intrusions
Quaternary Alluvium

330'0"N
330'0"N
Pinterang Formation
Tamako Group
Batunderang Group Binebase (65 m)
Malisang Group Bawone (86 m)
Binebase Group
Taware Group

Malisang (295 m)

Taware (306 m)
325'0"N

325'0"N
0 4
Km

12535'0"E 12540'0"E
Fig. 2. The geology of south Sangihe Island and the location of the Bawone and Binebase deposits, modified from Garwin
(1990).

Soil A BOD42 B
Pinterang Formation BOD1
Oxide zone
BID27 BOD3
Binebase Group
BID63 BID72
Andesitic crystal tuff BID69
BID16 BID56 Soil
BID60
BID57 Pinterang Formation
BID15 BID11 BID66 Oxide zone/weathered
Binebase Group
Porphyritic Intrusion
Binebase Group
Andesite porphyry
Andesitic tuff
Dacite

100 m

25 m

100 m 25 m

Fig. 3. Representative lithological cross sections (A) through the Binebase deposit (NW-SE) and (B) through the Bawone
deposit (SW-NE) based on drill holes. The black lines show the locations of the drill holes. Most primary features have been
destroyed in the oxide zone but the original nature of the rock is assumed to have been the same as that of the underlying
bedrock.
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1709

A B

Pl Mag

Bt

Hbl

C
D D
Kao + Qtz + Al

Py I
Py I + Qtz

Py II

Fig. 4. Photographs of the host rocks of the Bawone and Binebase deposits (cut slabs and core). (A) Argillically altered
flow-banded dacite, the basal unit of the Binebase Group at Bawone, showing phenocrysts and a matrix replaced by kaolinite
and quartz. (B) Plagioclase (Pl)-biotite (Bt), hornblende (Hbl)-magnetite (Mag) andesite porphyry. (C) Vein of py II (brassy)
with fragments of py I (black) in a matrix of vuggy py I with py II occasionally lining vugs. (D) Advanced argillic (alunite =
Al, pyrite = Py, quartz = Qtz) alteration of fragments (light gray) in a matrix of fine-grained quartz and pyrite (dark gray). The
scale bar is 1 cm.

Pinterang Formation (poorly consolidated volcaniclastic sand-
stones and siltstones, organic-rich siltstones, and calcareous
mudstones) ranges up to 100 m thick and is separated from
the Binebase Group rocks by a thin (several-cm) oxide layer,
possibly a paleosol. The Pinterang Formation is overlain by an
oxidized soil that is up to 4 m thick.
Hydrothermal Alteration
In hand sample, the altered rocks vary from dark gray to
white in color, depending on the pyrite content, and from very
hard (quartz dominated; Fig. 4C) to powdery (clay dominated;
Fig. 4D) in drill core; the latter variation commonly occurs
over intervals of tens of centimeters to tens of meters. Altera-
tion intensity is variable. In some samples, the primary min-
erals have been completely replaced by secondary minerals,
destroying all primary textures, whereas in other samples, vol-
canic textures are preserved (Figs. 4D, 5A). The pervasively
altered samples comprise very fine-grained, clay particles
intergrown with fine-grained, generally anhedral, equigranu-
lar quartz and pyrite (Fig. 5A, B, D). Less intensely altered
samples commonly retain porphyritic textures, although the
phenocrysts have generally been replaced by kaolinite or
dickite. The most intensely altered samples consist of a fine-
grained mosaic of intergrown anhedral quartz, clay minerals
(kaolinite, dickite, pyrophyllite), sulfate minerals (alunite,
natroalunite) and pyrite. Hereafter, the fine-grained, anhe-
dral pyrite intergrown with clay minerals and quartz will be
referred to as pyrite I (py I).
Alteration created secondary porosity in the form of mm-
scale vugs (formed by the dissolution of phenocrysts) and
irregular, cm-scale cavities. The latter are commonly infilled
by alunite, and/or pyrophyllite, and/or drusy pyrite, and/or
chalcocite, and/or barite crystals (Figs. 4C, 5C, D).
Alteration is most intense (few primary textures preserved)
along subvertical fluid conduits that fan out horizontally at
lithologic contacts (Fig. 6A). Distal to the mineralized zones,
alteration is generally less intense, primary volcanic textures
and minerals are preserved, and the rock is more competent.
The matrix of the rock is preferentially altered and has a grey
to light-brown color.
1710 KING ET AL.

A B

Qtz
Py I
Al

Py I

C D

Py II
Py II

Cc

Brt
Py I + Qtz

E F
Py II
Sp El

Ccp

Apy

En Brt

Brt

Fig. 5. Photomicrographs showing the textures of the altered and mineralized rocks in reflected light (10 m scale bar).
(A) Amphibole phenocrysts replaced by py I in a groundmass replaced by fine-grained quartz, py I, kaolinite dickite.
(B) Fine-grained association of quartz, alunite and py I, typical of argillic alteration. (C) Massive py II and a vug containing
rare chalcocite (Cc) and barite (Brt) crystals. (D) Growth zoned drusy crystals of py II infilling a vug in andesite altered to
fine-grained py I and quartz (blue-grey is epoxy). (E) Fractured and zoned crystals of massive py II cut by a barite-enargite
(En) vein. (F) Sphalerite (Sp), arsenopyrite (Apy), chalcopyrite (Ccp) and electrum (El) crystals in a coarse-grained barite
matrix. The earliest stage of gold-silver mineralization is manifest by fine-grained py I associated with equally fine-grained
quartz and clay minerals (A, B). This was followed by a second generation of Au-Ag mineralization in the form of coarse-
grained py II (C, D, E). The third generation of Au-Ag mineralization is represented by barite-enargite (En) veins that com-
monly contain brecciated py II crystals (E). Rare cm-scale veins of barite containing electrum, sphalerite, arsenopyrite and
chalcopyrite that cut py I and py II mineralization constitute a fourth generation of Au-Ag mineralization observed only in
the Binebase deposit.
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1711

Supergene zone A Overburden and B

IA I (qtz + py + kao dck) Pinterang Fm.
AA I (qtz+py+prlnal)
IA II (qtz + ill py) BID 27 AA II (qtz+pynaldckkao) BOD 42
BID 63 BID 72 AA III (qtz+py+kaoal)
BID 69 IA I (qtz+kaopy) BOD 1 BOD 3
BID 16 BID 56 BID 60
BID 11 BID 57 BID 66
BID 15

100 m

25 m

100 m
25 m

Fig. 6. Representative cross section (A) NW-SE through the Binebase (BID) deposit and (B) SW-NE through the Bawone
(BOD) deposit showing the distribution of hypogene alteration facies established using X-ray diffraction analysis and core log-
ging. The locations of drill holes are indicated by the black lines. Abbreviations: alunite = al, dickite = dck, illite = ill, kaolinite
= kao, natroalunite = nal, pyrite = py, pyrophyllite = prl, quartz = qtz.

Two main alteration facies have been recognized, advanced
argillic and intermediate argillic, based on the mineral asso-
ciations of quartz + pyrite + pyrophyllite + natroalunite +
alunite + dickite + kaolinite and quartz + pyrite + kaolinite
+ dickite + illite, respectively. Quartz and pyrite are nearly
ubiquitous alteration phases, but the proportions of these
minerals and those of the accompanying phases, which were
used to define the alteration facies, vary significantly. In prin-
ciple, the advanced argillic and intermediate argillic facies
can be subdivided into subfacies with fewer minerals that
reflect different physicochemical conditions. For example,
at quartz saturation (the case here), pyrophyllite is stable
at higher temperature than dickite, and dickite is stable at
higher temperature than kaolinite (Hemley et al., 1969; Stof-
fregen and Alpers, 1987; Stoffregen and Cygan, 1990). Simi-
larly, natroalunite is stable at higher temperature than alunite
provided that the K/Na ratio of the fluid is relatively con-
stant (Stoffregen and Cygan, 1990). Finally, illite is stable at
higher pH than kaolinite, dickite, and pyrophyllite. In view of
these relationships, mapping, X-ray diffraction and infrared
spectroscopy (TerraSpec) evidence were used to subdivide
the advanced argillic association into a high temperature,
lower pH subfacies (advanced argillic I) of quartz + pyrite
+ pyrophyllite natroalunite, an intermediate temperature,
low pH, subfacies (advanced argillic II) of quartz + pyrite
natroalunite dickite kaolinite, and a low temperature,
low pH subfacies (advanced argillic III) of quartz + pyrite +
kaolinite + alunite (see Discussion; Temperature and pH) .
The intermediate argillic association was subdivided into a
low pH subfacies (intermediate argillic I) of quartz + kaolin-
ite pyrite dickite and a higher pH subfacies of quartz +
illite pyrite (intermediate argillic II).
The alteration facies are zoned, both vertically and horizon-
tally, and the intensity of alteration decreases with depth and
laterally away from apparent fluid conduits. At Bawone, the
upper part of the deposit is dominated by advanced argillic
alteration. However, at Binebase the upper part of the deposit
is intensely oxidized due to weathering and supergene pro-
cesses, and advanced argillic alteration is not observed (Fig.
6B). The Bawone deposit was protected from the effects of
weathering by the overlying Pinterang Formation (Figs. 3A,
6A). Advanced argillic alteration in the Bawone deposit ter-
minates abruptly at the contact with the overlying Pinterang
Formation, although a thin (cm-scale) oxide layer is present
at this contact. This layer indicates that the uppermost part
of the deposit was previously oxidized, and perhaps partially
removed by erosion, prior to deposition of the Pinterang
Formation.
Advanced argillic alteration at Bawone forms subhorizon-
tal zones just below the Pinterang Formation and about 75 m
lower stratigraphically, at the contact with a zone of andesitic
tuff (Figs. 3A, 6A). Both zones appear to flare outward from
a steeply inclined advanced argillic alteration zone that likely
represents the conduit for the altering fluids. The upper parts
of the subhorizontal alteration zones and the central part of
the inclined alteration zone are dominated by the high tem-
perature, advanced argillic I alteration subfacies. Below the
subhorizontal advanced argillic I alteration zones and adjacent
to the inclined advanced argillic I alteration zone, this subfa-
cies changes to advanced argillic II alteration and distally to
intermediate argillic I alteration. Low temperature advanced
argillic III alteration is present only at depth and locally may
develop directly below advanced argillic I alteration without
intervening advanced argillic II alteration (Fig. 6A). Given its
distribution, the advanced argillic III alteration may be related
to the unmineralized hypabyssal porphyry intrusion (Fig. 3A).
At Binebase, the zone of supergene oxidation passes directly
into intermediate argillic alteration. The latter comprises a
1712 KING ET AL.

blanket of intermediate argillic I alteration with several sub-
vertical roots that pass downward and laterally into intermedi-
ate argillic II alteration (Fig. 6B).
Hypogene Mineralization
Neither free gold nor electrum is observed in the hypo-
gene zone of the Bawone deposit and very rarely in the Bine-
base deposit. Instead, the hypogene gold in both deposits is
invisible, i.e., hosted almost exclusively by pyrite. Silver is
also invisible, hosted by both pyrite and enargite, and cop-
per occurs mainly in pyrite and enargite (volumetrically much
less abundant than pyrite), as well as rare grains of chalcocite
in vugs (Fig. 5C). Gold (-silver) mineralization occurred in
three stages: (1) an early stage of disseminated Au-Agbear-
ing pyrite during advanced argillic and intermediate argillic
(minor) alteration (pyrite I; Figs. 5A, B, D, 7B), (2) an inter-
mediate stage characterized by multiple generations of lenses,
veins and breccias containing subhedral to euhedral, drusy,
and massive to semimassive Au-Ag-Cu-bearing pyrite II (py
II; Figs. 5C-E, 7A, C), and (3) late-stage barite-enargite-
pyrite II veins (Figs. 5C, E, 7C). At Binebase, there is a fourth
stage in the form of rare barite, base-metal sulfide, and elec-
trum veins (Figs. 5F, 7D).
The earliest gold occurs in py I, which is an ubiquitous
alteration product intergrown with quartz and hydrous alu-
mina-rich phases and occurs as small (230 m diameter),
isotropic, anhedral grains, accounting for 2 to 20 vol % of the
rock; the gold grade corresponds to the proportion of pyrite
in the rock. As part of the advanced argillic alteration and to
a much lesser extent the intermediate argillic alteration, this
stage of mineralization was focused along vertical structures
and spread out laterally near the top of the deposit where
fluids were confined by less-permeable strata or paleo-water
tables. At the Bawone deposit, advanced argillic I alteration is
also concentrated in a subhorizontal zone below an imperme-
able andesitic tuff about 75 m below the base of the Pinterang
formation (Figs. 3A, 6A).
The second stage of gold mineralization is characterized
by veins, veinlets, blebs, breccias, lenses, and massive bodies
of coarse-grained, brassy py II, which overprinted advanced
argillic alteration and intermediate argillic I alteration,
exploited the secondary porosity created during alteration,

A B

Qtz + Al

Py II

Py I

Brt

C D

Brt
Brt

Py II Ccp

El
En

Fig. 7. BSE images of auriferous pyrite, py I and py II, and later generations of mineralization. The scale bar is 100 m
unless otherwise indicated. (A) Fragment of drusy py II exhibiting strong growth zoning in a barite (Brt) vein. (B) Dissemi-
nated PyI in a matrix of quartz (Qtz) and alunite (Al). (C) Barite-enargite (En) vein with m-scale, homogeneous subhedral-
euhedral enargite crystals intergrown with mm-scale subhedral barite surrounding fragments of py II. (D) Electrum (El)
grain surrounded by subhedral chalcopyrite (Ccp), barite (Brt), and sphalerite (Sp) (10 m scale bar).
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1713
and crosscut fine-grained py I. As is the case with py I, the gold
grade is controlled by the abundance of py II, which forms 3
to 90 vol % of the rock, although py II is variably enriched
in gold. In thin section, py II is coarse grained (0.12 mm
long), massive or net-textured, and ranges in morphology
from anhedral to euhedral (Figs. 5A, D, 7A, C). Botryoidal
and drusy textures also are common, and many crystals are
complexly zoned (Fig. 7A). In reflected light, some pyrite II
crystals have an anhedral blue anisotropic core, which chemi-
cally is indistinguishable from the rest of the crystal (Fig. 5E).
Rare, isolated anhedral chalcocite crystals (3 mm in diameter)
are observed locally in vugs rimmed by drusy py II (Fig. 5C).
Growth zoning of py II crystals on the m-scale is common,
particularly in drusy crystals, which show a relatively stable
growth history recorded by continuous growth surfaces. It is
also observed in massive pyrite composed of crystals that grew
separately and were later annealed, resorbed, brecciated and
overgrown (Fig. 7A).
Pyrite I and II are commonly crosscut by barite-enargite
veins (0.25 cm wide). These veins appear to have re-opened
earlier py II veins, as they are commonly lined by fractured
py II crystals, and py II commonly occurs as brecciated frag-
ments in the veins, indicating that they were incorporated
mechanically during barite-enargite vein formation (Figs. 5E,
7C). In the centers of these veins, intact, blue-purple, iso-
tropic subhedral to euhedral (1060 m long) rectangular or
bladed enargite crystals and fractured py II fragments are sus-
pended in a matrix of coarse-grained, subhedral to drusy bar-
ite crystals. Texturally, the enargite is homogeneous, unzoned,
and appears to have coprecipitated with barite (Figs. 5E, 7C).
Barite and enargite also commonly occur in vugs. The den-
sity of barite-enargite veins (the proportion of veins relative to
host-rock) is variable but generally correlates with that of py
II veins, suggesting that both vein stages exploited the same
structures.
Although electrum is volumetrically insignificant, the gold
grade of these rare base-metal sulfide-electrum-barite veins is
20 times higher than that of the auriferous-pyrite dominated
zones of the deposit. In the Binebase deposit, these veins
range from 0.5 to 8 cm in width and are observed in the center
of the hypogene mineralization. This rare vein association was
only observed at Binebase, where it crosscuts all other min-
eralization stages, except for the enargite-barite veins. The
relative timing of these two late vein generations therefore
cannot be established. Mineralogically, the base metal sulfide-
electrum-barite veins comprise small subhedral to euhedral
crystals of spalerite, arsenopyrite, chalcopyrite, galena, pyrite,
electrum and Sb-sulfides that occur as aggregates in a coarse-
grained matrix of euhedral barite crystals (Figs. 5F, 7D).
Distribution of gold-silver mineralization
Gold and silver grades in the Bawone deposit are highest
just below the contact with the Pinterang Formation, decrease
with depth and then increase again just below a thin tuffa-
ceous andesite unit in a zone of advanced argillic I alteration
(Figs. 3A, 6A, 8). Metal concentrations correlate with the pro-
portion of py II, pyrite breccias and/or barite-enargite veins. A
similar geometry is inferred for the hypogene mineralization
at Binebase. However, because the upper part of the Bine-
base deposit was subjected to supergene oxidation, we infer
Au Ag Cu As Sb
40
Depth (m)
60
80
1000
2000
3000
1000
10
20
200
5000
50
250
Concentration (ppm)
Fig. 8. The concentrations of Au, Ag, Cu, As, and Sb (ppm) in a drill hole
from the Bawone deposit (BOD1); Pb and Zn concentrations were below the
detection limit and Mo and Ba were not analyzed.
that any high-grade hypogene mineralization in this deposit
was at the level of the supergene zone. This interpretation is
consistent with the observation that the remaining hypogene
mineralization has a much lower concentration of gold and
silver than the Bawone deposit (Table 1).
Assay data for 1m intervals of drill core for Bawone and
Binebase, made available by East Asia Minerals Corp., show
that there is a strong spatial correlation between Au and Ag
concentrations, and also of As with Cu, and a weaker correla-
tion of Cu and As with Sb (and Pb and Zn, not shown) concen-
trations (Fig. 8). Molybdenum concentrations are below the
2 ppm detection limit. The distribution of Au, Ag, As, and Cu
is vertically zoned, with the highest concentrations of these
elements occurring at the top of the hypogene zone. The
highest Au value coincides with high Cu, As, and Sb values
reflecting the presence of barite-enargite veins. This relation-
ship, however, does not hold for the second highest Au value
(Fig. 8). Indeed, most of the higher Cu, As, and Sb values are
accompanied by only minor to negligible enrichment in Au.
We interpret these metal distributions to indicate that barite-
enargite veins contributed only minor amounts of gold and
locally re-opened veins of py II to produce the coincident Au,
Cu, As, and Sb peaks.
Mass Changes During Alteration
Fifty-four drill core samples of the different lithological
units and alteration types were analyzed for major, trace, and
rare earth elements (REE) by inductively coupled plasma-
mass spectrometry (ICP-MS), and for gold by instrumental
neutron activation analysis (INAA), by Actlabs in Vancouver
(Table 2). The compositions of the altered rocks were com-
pared to those of the least-altered rocks to evaluate the gains
and losses of elements during alteration. In order to assess
these mass changes, it was first necessary to identify poten-
tially immobile elements that could be used to normalize
1714 KING ET AL.

Table 2. Major Element Compositions from Drill Core and Hand

IA I IA I IA I IA I IA I IA I IA I/BaEn IA I/BaEn
d.l. BOD1-134.5 BOD3-158.8 BOD3-85.4 BOD3-74.1 BOD3-106.4 BOD3-83.3 BOD3-99.2 BOD3-96.8

SiO2 (%) 0.01 55.4 64.68 8.2 58.66 51.41 48.19 19.69 85.78
TiO2 (%) 0.001 0.546 0.547 0.036 0.528 0.484 0.541 0.185 0.501
Al2O3 (%) 0.01 19.8 14.36 0.33 9.13 2.91 2.85 0.3 2.34
Fe2O3 (%) 0.01 4.26 8.38 35.08 13.45 27.98 30.09 40.04 6.14
MnO (%) 0.001 0.008 0.009 0.005 0.006 0.007 0.006 0.006 0.006
MgO (%) 0.01 0.75 0.02 0.02 0.01 0.02 0.02 0.02 0.02
CaO (%) 0.01 0.78 0.03 0.04 0.06 0.05 0.04 0.03 0.06
Na2O (%) 0.01 0.35 0.05 0.06 0.56 0.06 0.07 0.04 0.04
K2O (%) 0.01 2.17 0.08 0.02 0.69 0.02 0.04 0.05 0.04
P2O5 (%) 0.01 0.1 0.08 b.d. 0.06 0.04 0.05 b.d. 0.08
Total (%) 0.01 98.53 100.1 66.75 99.83 100.3 100.3 82.61 99.49
LOI (%) 14.36 11.82 23.02 16.69 17.33 18.38 22.25 4.49

La (ppm) 0.05 15.8 8.91 11 5.52 5.38 6.03 8.38 13.9

Ce (ppm) 0.05 32.1 16.5 7.29 8.67 8.77 10 8.03 26
Pr (ppm) 0.01 3.83 1.94 0.49 0.96 1.17 1.33 0.79 2.97
Nd (ppm) 0.05 14 8 1.21 3.95 4.15 4.67 2.55 12.1
Sm (ppm) 0.01 2.66 1.64 0.34 0.81 0.78 0.84 0.61 2.31
Eu (ppm) 0.005 0.791 0.596 b.d. 0.197 0.155 0.185 0.472 0.649
Gd (ppm) 0.01 2.18 1.2 1.51 0.61 0.57 0.58 1.02 1.49
Tb (ppm) 0.01 0.36 0.14 0.04 0.13 0.08 0.09 0.1 0.16
Dy (ppm) 0.01 2.19 0.72 0.09 0.76 0.42 0.44 0.28 0.89
Ho (ppm) 0.01 0.45 0.15 0.02 0.19 0.09 0.08 0.07 0.16
Er (ppm) 0.01 1.41 0.52 0.09 0.75 0.43 0.39 0.23 0.73
Tm (ppm) 0.005 0.237 0.103 0.014 0.157 0.08 0.067 0.042 0.146
Y (ppm) 0.5 12.8 4 1.3 5.6 3 2.8 2.5 5.2
Yb (ppm) 0.01 1.78 0.72 0.08 1.02 0.61 0.52 0.31 1.03
Lu (ppm) 0.002 0.31 0.119 0.012 0.173 0.101 0.091 0.053 0.174

Ag (ppm) 0.5 b.d. 0.5 11.5 b.d. 0.8 0.7 3.9 b.d.
As (ppm) 5 487 64 1,350 124 66 22 1,180 19
Au (ppb) 2 b.d. 306 1,500 751 1,110 1,190 2,000 723
Ba (ppm) 3 327 129 178,900 6,822 3,050 4,094 98,830 212
Bi (ppm) 0.1 b.d. 8.4 26.7 0.6 1 1.4 3.6 5
Co (ppm) 1 13 10 47 23 108 112 104 18
Cr (ppm) 20 b.d. b.d. b.d. b.d. 30 30 60 b.d.
Cs (ppm) 0.1 0.5 b.d. 0.2 b.d. 0.2 b.d. 0.3 b.d.
Cu (ppm) 10 40 170 4,550 1,780 490 2,590 4,540 270
Ga (ppm) 1 18 28 4 7 10 7 3 8
Ge (ppm) 0.5 1.8 2.8 0.6 b.d. b.d. b.d. 0.5 1.2
Hf (ppm) 0.1 2.2 1.8 0.2 2.4 2 1.6 0.8 3.1
In (ppm) 0.1 b.d. b.d. 0.4 b.d. b.d. b.d. 0.2 0.1
Mo (ppm) 2 b.d. 5 28 4 12 6 14 13
Nb (ppm) 0.2 6.4 2.2 1.1 2.7 2.6 2.7 1.3 2.3
Ni (ppm) 20 b.d. b.d. b.d. b.d. 30 30 40 b.d.
Pb (ppm) 5 8 49 174 12 38 25 57 18
Rb (ppm) 1 36 b.d. b.d. 1 3 b.d. 2 b.d.
Sb (ppm) 0.2 b.d. 2.8 112 10 2.5 b.d. 70.5 4.8
Sc (ppm) 1 11 13 b.d. 9 7 5 3 10
Sn (ppm) 1 b.d. b.d. 15 b.d. 1 b.d. 5 b.d.
Sr (ppm) 2 248 702 1,484 477 298 310 772 515
Ta (ppm) 0.01 0.35 0.18 0.06 0.18 0.2 0.15 0.08 0.21
Th (ppm) 0.05 2.79 1.53 0.5 1.51 1.19 1.96 0.5 2.05
Tl (ppm) 0.05 2.28 1.34 4.56 0.68 1.31 0.48 0.16 0.36
U (ppm) 0.01 1.02 0.59 0.14 0.47 0.87 0.55 0.38 1.68
V (ppm) 5 128 158 8 80 46 39 11 33
W (ppm) 0.5 b.d. 4.1 19.2 4 4.4 9.9 2.6 6.4
Zn (ppm) 30 1,060 140 b.d. 1,030 360 100 b.d. 320
Zr (ppm) 1 81 78 6 86 75 58 30 144
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1715

Samples from the Bawone (BOD) and Binebase (BID) Deposits

Porphyry

AA I AA I AA I AA I AA I AA I AA I/BaEn AA II AA II AA II AA II
BOD3-49.8 BOD3-55 BOD3-85.5 BOD3-48.6 BOD3-100 BOD3-78.7 BOD3-113.7 BOD42-62.3 BOD3-125.4 BOD3-46.9 BOD3-34.4

56.29 43.28 42.65 48.8 29.11 8.74 63.84 49.79 51.37 53.65 20.65
0.53 0.373 0.492 0.492 0.377 0.22 0.43 0.46 0.474 0.603 0.684
13.22 6.17 3.28 10.38 0.26 1.25 7.59 10.59 11.21 12.29 19.21
8.23 15.17 32.38 17.76 46.64 59.44 10.06 16.73 12.06 12.72 30.85
0.009 0.01 0.007 0.007 0.009 0.014 0.005 0.01 0.003 0.008 0.012
0.01 0.01 0.02 b.d. 0.02 0.03 0.01 0.02 b.d. 0.02 0.03
0.08 0.05 0.04 0.06 0.04 0.05 0.06 0.06 0.07 0.12 0.07
0.76 0.42 0.15 0.48 0.05 0.06 0.84 0.38 0.92 0.71 0.17
1.35 0.47 0.15 0.71 0.03 0.02 0.81 0.36 1.21 0.71 0.18
0.12 0.04 0.04 0.1 0.02 0.02 0.08 0.12 0.08 0.2 0.22
99.58 84.1 99.82 99.86 100.1 100.8 99.15 97.66 97.92 100 99.21
18.98 18.11 20.61 21.06 23.58 31.01 15.42 19.13 20.52 19 27.12

8.79 9.01 4.95 7.79 2.93 2.58 6.53 8.76 7.62 10.8 8.29
16.3 14.8 8.42 14 4.48 3.36 11.5 17 13.9 20.5 15.5
1.92 1.96 1.14 1.62 0.57 0.41 1.63 2.08 1.61 2.43 2.06
7.94 7.17 3.91 6.71 2 1.35 5.76 8.38 6.79 10.4 7.02
1.49 1.56 0.75 1.36 0.37 0.26 1.13 1.8 1.43 2.33 1.25
0.442 0.349 0.176 0.377 0.086 0.058 0.318 0.453 0.425 0.689 0.319
1.13 1.55 0.53 0.98 0.27 0.2 0.82 1.23 1.15 2.15 0.87
0.18 0.16 0.07 0.16 0.04 0.03 0.14 0.15 0.15 0.31 0.13
1.04 0.69 0.39 0.82 0.26 0.16 0.75 0.77 0.85 1.69 0.71
0.25 0.15 0.08 0.19 0.08 0.03 0.15 0.16 0.17 0.36 0.16
0.86 0.56 0.35 0.73 0.3 0.14 0.66 0.56 0.62 1.21 0.59
0.174 0.101 0.065 0.148 0.057 0.026 0.134 0.104 0.122 0.237 0.115
6.6 5.2 2.5 5.5 2.1 1.1 4.5 4 5.2 10.4 4
1.16 0.78 0.45 1.04 0.46 0.17 0.9 0.84 0.84 1.64 0.83
0.189 0.13 0.085 0.177 0.075 0.029 0.154 0.162 0.137 0.274 0.138

b.d. 0.6 b.d. b.d. 2.3 2.2 0.6 1 b.d. b.d. 1.9
38 314 14 20 167 36 8 74 28 56 141
166 666 1,470 580 1,710 1,690 287 487 300 396 1,390
779 95,970 1,990 1,344 1,473 511 370 1,865 688 1,458 1,736
b.d. b.d. 0.3 b.d. 5.6 1.7 0.1 3.2 1.1 0.2 0.7
10 22 115 10 157 188 25 9 28 12 28
20 50 b.d. b.d. 20 20 30 b.d. b.d. b.d. 30
b.d. 0.1 b.d. b.d. 0.2 b.d. b.d. 0.3 b.d. b.d. 0.2
230 1,550 1,450 1,010 2,870 2,010 490 410 350 710 1,460
18 19 7 32 1 3 11 10 10 48 58
0.8 b.d. b.d. b.d. b.d. b.d. b.d. 0.9 0.6 b.d. b.d.
2.5 1.9 1.5 2 2.4 0.6 2.2 2 1.7 2.8 3.3
b.d. 0.2 b.d. b.d. 0.1 b.d. 0.1 b.d. 0.2 b.d. b.d.
b.d. 3 4 b.d. 21 29 3 2 2 4 9
2.6 2 2.6 2.2 1.9 1 2.1 2.1 2 3 2.9
b.d. b.d. 30 b.d. 40 60 b.d. b.d. b.d. b.d. 30
23 b.d. 47 62 84 15 14 354 14 30 46
b.d. 2 1 2 b.d. 1 b.d. 2 b.d. 1 1
1 6.5 b.d. b.d. 11.5 9.3 0.8 2.1 1.3 1.9 8.7
13 8 6 13 5 2 8 11 10 13 28
b.d. b.d. b.d. b.d. 2 b.d. b.d. 5 1 b.d. 4
669 564 310 592 60 83 455 641 532 931 1,717
0.2 0.14 0.14 0.12 0.14 0.06 0.14 0.15 0.15 0.19 0.21
1.61 1.3 1.08 1.38 0.72 0.31 1.05 1.24 1.25 1.78 2.41
0.72 0.33 0.75 4.03 0.61 0.82 0.6 2.7 1.15 2.52 4.41
0.51 0.45 0.79 0.68 0.77 0.35 0.56 0.57 0.54 0.63 1.25
132 73 42 148 21 32 88 94 109 174 259
3.3 1.6 20.3 0.8 1.7 6.1 46 0.5 6.4 7.9 2.2
430 260 360 460 80 50 270 120 500 1,070 290
95 68 59 70 87 18 69 65 75 98 99
1716 KING ET AL.

Table 2.

Porphyry

AA II AA II AA II AA II AA II AA II/BaEn AA III AA III AA III

BOD3-27.3 BOD3-79.5 BOD3-132.9 BOD3-41 BOD3-65.4 BOD3-65.8 BOD1-115 BOD1-124.9 BOD3-122.6

SiO2 (%) 7.08 65.52 59.09 39.59 56.91 17.64 59.72 61.62 39.9
TiO2 (%) 1.029 0.485 0.624 0.567 0.431 0.082 0.61 0.773 0.298
Al2O3 (%) 24.05 6.24 16.11 3.97 5.67 0.81 16.49 16.96 3.6
Fe2O3 (%) 19.86 14.38 7.88 32.64 18.17 31.27 7.98 5.73 32.09
MnO (%) 0.005 0.006 0.016 0.017 0.005 0.003 0.014 0.022 0.006
MgO (%) 0.03 b.d. 0.02 0.02 0.02 0.02 0.02 0.03 0.02
CaO (%) 0.12 0.05 0.08 0.07 0.08 0.04 0.11 0.1 0.04
Na2O (%) 1.22 0.05 0.23 0.16 0.43 0.06 0.07 0.14 0.31
K2O (%) 3.49 0.01 0.17 0.21 0.42 b.d. 0.03 0.03 0.29
P2O5 (%) 0.43 0.1 0.18 0.13 0.11 b.d. 0.21 0.11 0.04
Total (%) 98.86 98.17 98.46 99.66 100 77.25 98.39 99.98 98.21
LOI (%) 41.55 11.31 14.06 22.28 17.79 22.94 13.13 14.47 21.62

La (ppm) 16.8 8.97 11.4 10.4 8.48 6.84 12.1 9.83 3.87
Ce (ppm) 31.1 15.8 22.1 19.4 15.4 5.76 24.7 17.9 5.81
Pr (ppm) 3.44 2.12 2.59 2.68 2.08 0.53 2.84 2.1 0.71
Nd (ppm) 13.5 7.48 11.2 9.94 7.45 1.48 12 8.17 2.53
Sm (ppm) 2.4 1.34 2.45 2.19 1.4 0.35 2.44 1.56 0.53
Eu (ppm) 0.496 0.368 0.854 0.68 0.415 0.18 0.616 0.784 0.181
Gd (ppm) 1.53 0.88 1.96 1.77 1 0.76 1.59 1.19 0.49
Tb (ppm) 0.19 0.12 0.24 0.22 0.15 0.05 0.2 0.2 0.06
Dy (ppm) 0.95 0.65 1.23 1.14 0.77 0.17 1.04 1.25 0.28
Ho (ppm) 0.2 0.13 0.26 0.22 0.18 0.03 0.23 0.28 0.06
Er (ppm) 0.71 0.6 0.99 0.83 0.74 0.13 0.78 0.99 0.24
Tm (ppm) 0.14 0.123 0.184 0.144 0.151 0.022 0.142 0.18 0.044
Y (ppm) 5.3 4.2 6.9 6.6 5.1 1.7 6.2 7.1 1.8
Yb (ppm) 1.06 0.9 1.35 1.02 1.07 0.16 1.17 1.42 0.31
Lu (ppm) 0.183 0.15 0.218 0.164 0.174 0.028 0.23 0.28 0.055

Ag (ppm) 3.2 b.d. b.d. 1.6 2.8 3.7 b.d. 1.1 4.5
As (ppm) 55 106 104 442 311 2,500 60 125 242
Au (ppm)
1,240 717 345 1,310 1,080 1,280 273 130 1,760
Ba (ppm)
1,546 2,354 1,517 2,519 4,145 106,900 802 1,609 13,280
Bi (ppm) 1.2 b.d. 1.5 0.7 0.3 0.3 0.4 1 0.6
Co (ppm) 17 53 11 25 26 54 8 10 111
Cr (ppm) 40 30 20 30 30 30 30 20 30
Cs (ppm) 0.1 b.d. b.d. b.d. b.d. b.d. b.d. 0.3 0.3
Cu (ppm) 640 1,440 180 3,500 2,720 15,000 70 90 2,470
Ga (ppm) 184 13 18 24 19 6 13 19 8
Ge (ppm) b.d. 1.7 0.7 0.6 1.2 1.2 3.7 2.1 0.6
Hf (ppm) 4 3.6 2.1 2.1 2.2 0.3 2.4 2.8 1.1
In (ppm) 0.2 b.d. b.d. b.d. 0.1 0.2 b.d. b.d. 0.1
Mo (ppm) 12 18 15 11 5 21 58 49 38
Nb (ppm) 4.2 2.4 2.6 2.2 2.2 0.6 4.2 2.9 1.5
Ni (ppm) 20 20 b.d. 30 b.d. 20 b.d. b.d. 50
Pb (ppm) 149 b.d. 17 11 181 b.d. 74 120 28
Rb (ppm) 2 b.d. b.d. 1 1 1 b.d. b.d. 2
Sb (ppm) 8.8 34.4 2.1 35.6 24.6 120 4.5 10.4 27.7
Sc (ppm) 35 10 14 10 9 2 14 23 5
Sn (ppm) 5 3 b.d. 6 4 6 4 2 3
Sr (ppm)
2,840 682 1,060 655 571 1,109 685 295 411
Ta (ppm) 0.3 0.21 0.2 0.17 0.16 0.04 0.19 0.2 0.1
Th (ppm) 3.8 1.66 1.85 1.52 1.33 0.32 1.83 1.93 0.52
Tl (ppm) 3.81 0.93 1.53 0.49 4.59 0.07 1.21 3.13 0.4
U (ppm) 5.45 1.91 1.45 0.84 0.51 0.25 0.67 12.6 0.4
V (ppm) 410 69 157 91 59 17 120 199 36
W (ppm) 5.3 2.1 0.8 3.4 12.9 1.6 b.d. b.d. 2.5
Zn (ppm) 310 40 120 70 300 190 360 960 b.d.
Zr (ppm) 141 109 95 55 73 9 89 101 36

Note: The sample name indicates the drill hole number and the depth; samples are organized by lithology (porphyry or crystal tuff) and alteration-type
(IA = intermediate argillic, BaEn = barite-enargite veins, AA = advanced argillic; see text for description), identified based on petrographic observations,
bulk rock geochemistry and/or X-ray diffraction analyses
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1717

(Cont.)

Crystal Tuff

Least altered IA I IA I IA I IA II IA II IA II IA II IA II IA II IA II
PANTAI BID70-61.7 BID24-101 BID39-41.2 BID43-39.5 BID26-102.5 BID15-84.2 BID18-144.5 BID18-183.2 BID23-68.6 BID26-58

58.37 63.89 65.1 29.58 66.18 60.58 58.02 56.54 56.83 56.02 47.52
0.546 0.627 0.706 1.423 0.587 0.68 0.771 0.796 0.852 0.667 0.902
15.2 11.62 12.71 26.73 12.6 14.96 16.24 17.84 18.21 15.5 19.49
5.59 9.89 8.13 18.12 5.45 6.7 7.15 6.92 6.79 6.86 6.34
0.132 0.005 0.005 0.004 0.017 0.023 0.018 0.015 0.027 0.019 0.025
2.86 0.04 0.03 0.06 0.99 1.04 0.96 1.86 0.84 1.15 0.68
7.02 0.06 0.07 0.11 0.33 0.38 0.14 0.45 0.8 0.57 0.62
2.79 0.04 0.04 0.07 0.04 0.03 0.07 0.06 0.06 0.06 0.09
1.1 0.26 0.02 0.05 3.55 3.79 4.29 2.12 2.41 4.17 6.09
0.2 0.15 0.16 0.43 0.16 0.17 0.15 0.23 0.2 0.14 0.23
99.55 98.27 98.77 98.61 99.88 99.07 99.25 98.31 99.5 100.5 98.29
5.74 11.69 11.81 22.04 9.97 10.72 11.44 11.47 12.49 15.38 16.31

12.5 7.53 9.05 10.5 6.12 6.04 8.04 10.3 10.5 11.9 12.7
24.6 14.6 18.6 16.1 11.7 13.5 18.3 21.5 21.7 23.3 25.6
3.13 1.52 2.13 1.35 1.46 1.69 2.52 2.83 2.93 3.04 3.33
12.8 6.08 8.82 3.57 6.05 7.94 11.1 12.2 12.4 12.6 14.1
2.96 1.05 1.88 0.39 1.98 2.51 2.98 3.37 3.36 2.91 3.51
1.03 0.352 0.595 0.2 0.581 0.84 0.858 0.947 1.09 0.891 1.1
2.89 0.73 1.36 0.32 2.46 3.12 3.11 3.49 3.58 2.97 3.74
0.48 0.12 0.17 0.05 0.46 0.57 0.55 0.58 0.63 0.52 0.65
2.87 0.81 0.86 0.32 2.94 3.59 3.39 3.58 3.9 3.28 4.01
0.6 0.21 0.19 0.09 0.61 0.73 0.69 0.77 0.81 0.68 0.82
1.81 0.75 0.71 0.37 1.79 2.19 2.04 2.38 2.36 2.04 2.42
0.283 0.136 0.135 0.079 0.277 0.344 0.324 0.37 0.379 0.319 0.382
16.6 6.1 5.5 2.4 17.7 21.7 19.2 21.6 22.2 19.2 22.4
1.97 1.09 1.1 0.75 1.92 2.4 2.39 2.62 2.67 2.25 2.64
0.331 0.205 0.209 0.173 0.322 0.411 0.435 0.445 0.466 0.374 0.452

b.d. 0.6 b.d. 1.9 12.7 6.5 3.2 b.d. b.d. b.d. b.d.
b.d. 10 19 25 204 43 15 9 10 94 118
b.d. 57 66 391 317 27 54 21 13 b.d. b.d.
181 664 107 1,837 8,105 2,045 1,381 130 139 749 744
b.d. 0.4 0.5 4.2 b.d. b.d. b.d. b.d. b.d. b.d. b.d.
14 8 21 80 16 14 14 21 21 16 18
20 b.d. b.d. 200 20 b.d. b.d. 20 b.d. b.d. b.d.
1.5 b.d. b.d. 0.3 1.1 2 1.7 1.2 1.3 3 4.4
30 200 100 490 160 40 50 60 80 60 70
14 8 12 16 14 14 15 16 17 15 19
1.5 1.4 1.7 2 0.8 0.6 0.5 1.5 1.7 0.7 0.8
2.3 2.1 2.2 4.2 1.9 2.2 2.4 2.6 2.4 2 2.7
b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d. b.d.
b.d. b.d. 10 19 16 3 b.d. 3 b.d. b.d. b.d.
2.7 2.7 2.4 4.9 2.1 2.3 2.8 2.7 3.1 2.3 3.3
b.d. b.d. b.d. 30 b.d. b.d. b.d. b.d. b.d. b.d. b.d.
b.d. 15 25 55 1,080 44 54 10 23 30 b.d.
21 4 b.d. 3 79 82 93 42 52 80 105
b.d. 1.7 b.d. 4.1 12.7 9.7 1.8 b.d. 6.7 11.5 9.7
18 14 14 18 21 26 29 25 32 25 32
b.d. 3 3 11 2 b.d. b.d. b.d. b.d. b.d. b.d.
367 1,040 688 3,854 184 73 59 23 47 54 61
0.17 0.15 0.17 0.3 0.13 0.14 0.16 0.15 0.16 0.12 0.18
1.83 1.22 1.03 1.73 1.29 1.23 1.63 1.59 5.43 1.31 1.85
0.1 0.38 0.75 0.39 4.04 0.9 1.07 0.35 0.28 3.85 2.83
0.77 0.38 0.91 0.59 5.01 0.74 0.61 0.52 1.84 0.6 0.65
171 120 123 222 178 201 248 245 285 221 283
2.4 0.8 0.8 2.4 3.6 5.6 8.7 2 1.3 b.d. b.d.
60 b.d. 80 b.d. 3,450 320 300 70 320 350 470
89 72 80 130 67 76 82 92 85 70 100
1718 KING ET AL.
the compositions of the altered rocks to those of their least-
altered equivalents. This was done by making binary plots of
the data, and determining which element pairs were linearly
distributed, as such a distribution would correspond to a rela-
tively constant mass ratio for the elements and could indicate
that they were immobile (e.g., MacLean and Kranidiotis,
1987; Warren et al., 2007; Agrawal et al., 2008). These plots
show that concentrations of pairs of the following elements,
Zr, Ti, Nb, Hf, and Ta, form linear arrays. However, because
Ti, Nb, and Ta are not incorporated into the same minerals as
Zr and Hf (Ti, Nb, and Ta occur in rutile and Zr and Hf in zir-
con), the linear variation of Zr with Ti was taken as evidence of
immobility (Fig. 9). Assuming that Zr and Ti were immobile,
the concentration of these elements in each altered sample
was normalized to their concentration in a least altered sam-
ple for each rock type using the method of Grant (1986). To
scale the changes for each element proportionally so that the
mass change for an unaltered rock is zero, the concentration
of each element was multiplied by the ratio of the immobile
element concentrations (Zr and Ti) in the fresh rock to that in
the altered rock. Values less than unity indicate losses of ele-
ments relative to their concentrations in the unaltered rock,
and values greater than unity indicate gains (Fig. 10).
Three distinct patterns of relative gains and losses are
observed for advanced argillic alteration (the small size
of the data set precluded separate treatment of advanced
argillic I, II, and III alteration) and intermediate argillic I
(Bawone and Binebase) and II (Binebase) alteration (Fig.
10). As expected from the mineralogy, the advanced argillic
and intermediate argillic I samples were strongly depleted
in the major elements, Mn, Mg, Ca, and Na, moderately
depleted or unchanged in K, weakly depleted in Al and P,
and relatively undepleted in Si. Somewhat unexpectedly,
however, these samples appear to be enriched in Fe. As py II
1.6
BOD
BID
Least Altered
1.2
TiO2 (wt. %)
0.8
0.4
0 The compositions of pyrite and enargite were analyzed
40 80 120 160
Zr (ppm)
Fig. 9. Plot showing the distribution of TiO2 and Zr in variably altered
rocks of the Bawone (BOD) and Binebase (BID) deposits. See the text for an
interpretation of the significance of these data. The Pearson correlation coef-
ficient for both deposits (r) = 0.80.
commonly occurs in the form of blebs that filled pores cre-
ated by advanced argillic and intermediate argillic I altera-
tion (and thus postdated this alteration), we suspect that most
if not all of the apparent enrichment in Fe is an artifact due
to unavoidable inclusion of py II in the samples analyzed to
represent this alteration. The intermediate argillic altera-
tion II samples were also depleted in Mn, Mg, Ca, P, and
particularly in Na, but appear to have conserved Al, Si, and
Fe. In contrast to advanced argillic and intermediate argillic
I alteration, these samples are strongly enriched in K. The
overall distribution of trace metals and semimetals (Au, Cu,
As, Ag, Cr, Co, Ni, Zn, Mo, Sn, Sb, Tl, W, and Pb) indicates
that they were added during all stages of alteration, albeit in
variable quantities (it should be noted that because of the py
II problem referred to above, the additions of Cu and Au for
advanced argillic and intermediate argillic I alteration may be
overestimated). However, whereas Rb and Cs were depleted
(or unchanged) and Sr added during advanced argillic and
intermediate argillic I alteration, the opposite occurred dur-
ing intermediate argillic II alteration, i.e., Sr was depleted
and Rb and Cs enriched (Fig. 10).
Both the advanced argillic and intermediate argillic I
altered rocks underwent substantial mass losses in all REEs,
with the degree of depletion increasing progressively with
atomic number from La to Dy and then decreasing progres-
sively to Lu (Fig. 10). In contrast to advanced argillic and
intermediate argillic I alteration, the depletion of REEs dur-
ing intermediate argillic II alteration was greatest for La and
decreased with atomic number to Gd with the exception of
Eu. We consider that the very small additions of the heavy
rare earth elements (HREEs: Tb, Dy, Ho, Er, Tm, Yb, Lu)
reflect uncertainties in the mass transfer calculations and that
these elements were immobile during alteration. The prefer-
ential leaching of the light rare earth elements (LREEs: La,
Ce, Pr, Nd) during intermediate argillic II alteration is con-
sistent with results of experiments showing that the LREEs
are more mobile in chloride-bearing hydrothermal fluids than
the HREEs (Migdisov et al., 2009). The reason for the prefer-
ential depletion of Eu, Gd, Tb, Dy, and Ho during advanced
argillic and intermediate argillic I alteration is unclear. A pos-
sible explanation for this anomalous behavior is that depletion
of the LREEs was inhibited because of their incorporation
by sulfates and clay minerals (e.g., Miller et al., 1982; Hopf,
1993; Fulignati et al., 1999; Karakaya, 2009), and that with-
out this uptake they would have been more depleted, con-
sistent with the experimental predictions (Migdisov et al.,
2009). The differential behavior of the REEs reported above
supports observations from active hydrothermal systems that
REE chemistry offers a useful tool for distinguishing altera-
tion types (e.g., Michard, 1989; Fulignati et al., 1998, 1999;
Lottermoser, 1992; Lewis et al., 1997; Salan et al., 2011).
Ore mineral composition
using a combination of electron microprobe (EMP) and laser
ablation-inductively coupled plasma-mass spectrometry (LA-
ICP-MS). Quantitative electron microprobe analyses for Fe, S,
Cu, As, Sb, Co, Ni, Zn, Se, and Te were conducted on carbon-
coated samples at McGill University using a JEOL 8900 instru-
ment equipped with five wavelength dispersive spectrometers
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1719

10

Fe 2O 3
10 1000
Advanced Argillic

K2O
1 SiO 2 100
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

P2O 5
Al2O 3 1
0.1 10

MnO

MgO

Na2O
CaO
0.01 1

Rb
Sr
Sn
Sb

Pb
Zn

Ba
Au

As
Ag

Mo
Cu

Co

Bi
Ni

Cs
0.1
0.1

10
Intermediate Argillic I

1000
Fe 2O 3

10
100
1
SiO 2

Al2O 3

10 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
P2O 5

1
K2O

0.1
Na2O

1
MnO

Sn

Pb
Zn
Au

As
Ag

Mo

Sr
Cu

Co

Bi
Ni
Sb

Ba
0.01
0.1
CaO
MgO

Cs
Rb
0.1
10
Intermediate Argillic II

1000
10
K2O
Fe 2O 3

100
1
Al2O 3

10 La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
P2O 5
SiO 2

MnO

MgO

0.1
1
CaO

1
Rb

Cs
Ba
Pb
Zn
Sn
Sb
As

Mo

Bi
Co
Cu
Ag

Ni
Au

0.01
Na2O

0.1
Sr

0.1
Fig. 10. Relative gains and losses of elements for advanced argillically altered rocks from Bawone (red), intermediate
argillically altered I porphyry and crystal tuffs from Bawone and Binebase (yellow) and intermediate argillically II altered
(purple) crystal tuffs from Binebase, calculated using the method of Grant (1986). A value of 1 indicates immobility, less than
1 indicates a relative mass loss and greater than 1 indicates a relative mass gain.

(WDS). The operating conditions were an excitation potential The concentrations of Fe and S determined by EMP analysis
of 20 kV, a beam current of 50 nA and a spot-size of 2 m. were used as an internal standard for the LA-ICP-MS analy-
Analyses were standardized using pyrite (Fe, S), chalcopyrite ses of pyrite and enargite, respectively; GSE-IG, NIST 610
(Cu), AsNiCo (As, Ni, Co), CdTe (Cd, Te), sphalerite (Zn), and Po689 were used as external standards for the analysis of
stibnite (Sb), and AgSe (Se) supplied by CANMET. In addi- pyrite and GSE-1G, PS1 and JB5 for enargite.
tion to quantitative spot analyses, the EMP was also used to Pyrite: The following trace elements were detected in pyrite
produce element maps for Cu, Fe, Se, and As. The operating using the EMP: As, Co, Cu, Ni, Sb, Se, Te, and Zn (Table AI).
conditions were an excitation voltage of 20 kV, an operating LA-ICP-MS analyses yielded results for these elements simi-
current of 90 nA, and a beam diameter of 2m. The counting lar to those obtained with the EMP, and also detected Ag, Au,
time was 30 ms and the pixel size between 0.20 and 1.28 m. Bi, and Pb (Table A2). Copper is the principal trace element
The LA-ICP-MS analyses were conducted at the Geologi- in both generations of pyrite followed by Co, As, and Pb in py
cal Survey of Canada (pyrite) using a Photon Machines Nalyte I or by Co, Pb, and As in py II. Significantly, both generations
193 mm Excimer laser coupled to an Agilent Technologies of pyrite contain ppm levels of Au and Ag.
7700 Series ICP-MS, and at Universit de Chicoutimi (enar- Pyrite I crystals are compositionally unzoned (Fig. 7B) and
gite) using an Excimer Resolution M-50 (Resonetics) Laser have a wide inter- and intra-crystal range of Cu concentrations
coupled to an Agilent 7700x ICP-MS. The concentration of (0.021.5 wt %); the median Cu concentration is 0.44wt%
Fe57, S34, Cu65, Au197, Se77, Te125, Ag107, As75, and Sb121 was and the interquartile range (IQR) is 0.24 to 0.72 wt%. Cobalt
measured in both pyrite and enargite. Pyrite was also ana- concentration is significantly lower, with a median of 62.5ppm
lyzed for Co59, Ni60, Pb208, Zn66, and Bi209. The ablation pits (IQR = 15.6206 ppm). The median As concentration is
ranged from 14 to 54 m in diameter, and the counting time 43.9ppm (IQR = 15.683 ppm) and that of Pb is 59.6 ppm
was 100 s (30 s of background, 70 s of ablation) for pyrite (IQR = 9.579.7 ppm). As mentioned above, py I is aurifer-
and 90 s (30 s for background, 60 s of ablation) for enargite. ous and argentiferous. The median concentrations of Au and
1720 KING ET AL.

Ag are 1.0 ppm (IQR = 0.61.5 ppm) and 9.1 ppm (IQR =
1.129.6 ppm), respectively. Somewhat surprisingly, the con-
centrations of Se and Te, elements that with As are commonly
elevated in auriferous pyrite (e.g., Fleet et al., 1993; Reich et
al., 2005), are relatively low with median values of 9.6 ppm
(IQR = 6.141.1 ppm) and 3.8 ppm (IQR = 2.210 ppm),
respectively (Table 3).
The concentrations of all trace elements detected in py I
are higher in py II. The median concentration of Cu in py II
is 1.1wt % (IQR = 0.423.4 wt %) and the maximum concen-
tration 6.0 wt % (EMP Analysis, Table A1). The median con-
centration of Co in py II is 148 ppm (IQR = 48.4533ppm),
that of Pb is 69.9 ppm (IQR = 2.7557 ppm) and that of As is
118ppm (IQR = 16.1417 ppm). Unlike py I, the concentra-
tion of Te is relatively high, with a median value of 31.3ppm
(IQR = 11143 ppm) and that of Se is significant, with a
median of 25.9 ppm (IQR = 12.8102 ppm). The maximum Au
concentration in py II is 13.6 ppm, and the median, 2.4ppm
(IQR = 0.93.2 ppm), and the Ag maximum concentration is
1,273ppm, and the median, 3.7 ppm (IQR = 1.419 ppm).
Although the concentrations of the trace elements, on aver-
age, are lower in py I than in py II, the interelement asso-
ciations are similar. Binary plots (Fig. 11) and element maps
(Fig. 12) show that high Au concentrations in py I and II are
generally associated with elevated Cu, As, Ag, and Te, but Au
concentration may be high even if concentrations of these
elements are relatively low; this is especially true for Ag and
Te. Gold concentration is almost entirely independent of Se
concentration and shows no correlation with Co (or Ni, Sb,
Zn, Pb, and Bi, which are not shown) concentration. The
inter-element correlations for Ag are similar to those for gold,
except that Ag and Te are negatively correlated.
As mentioned above, py II is complexly compositionally
zoned and py I is unzoned. The backscatter electron images
indicate that this zonation reflects differences in concentra-
tions of trace elements, mainly of Cu (Fig. 7A). Compositional
zoning was confirmed by quantitative spot analyses of py II,
which show that there are appreciable compositional varia-
tions within single crystals.
Copper concentrations show the largest variation between
growth zones, the distribution of which reveals a complex
history of growth, overgrowth, and resorption (Fig. 12). In
addition to this growth zoning, sector zoning is evident in the
distribution of Cu (i.e., within single growth zones there are
sharply defined sectors of higher and lower Cu concentration
corresponding to differences in the crystal faces presented).
The nature of the sector zoning varies considerably with the
crystal habit and is most easily observed in drusy crystals (Fig.
12).
Based on electron microprobe analyses, the Cu concentra-
tion of individual zones can range from as low as 0.12 to as
high as 6.0 wt %, and vary inversely with the concentration
of Fe. In some zones, high Cu concentration is matched by
an elevated concentration of As, but in others, the concen-
trations of the two elements do not correlate positively (Fig.
12D, F). In addition to As, the concentrations of Ni, Se, and
Te were measured across the transects. All three elements
show increases in concentration with elevated Cu concen-
tration. Where the concentrations are consistently above the
detection limits, Ni, Se, and Te covary. In some pyrite crystals,
they covary with Cu, in others with As, and in some crystals,
with both.
Further insight into how trace elements in py II covary
is provided by LA-ICP-MS depth profiles through crystal
growth zones. Overall, the profiles show that zones enriched
in Cu are enriched in all trace elements, although the con-
centration of individual elements covaries with some, but not
all, of the elements analyzed. This is illustrated in Figure 13,
which shows the signal intensity (counts) for a suite of ele-
ments through three growth zones. From this figure, it is evi-
dent that the concentrations of Au, As, and Te covary (they all
decrease strongly in the central zone), that Cu, Co, Ni, Zn, Bi,

Table 3. Average Composition of Py I, Py II, and Enargite Based on LA-ICP-MS Analyses

Ag As Au Bi Co Cu Fe Ni Pb S Sb Se Te Zn
(ppm) (ppm) (ppm) (ppm) (ppm) (wt %) (wt %) (ppm) (ppm) (wt %) (ppm) (ppm) (ppm) (ppm)

Py I Mean 33.0 95.4 1.1 10.1 117.1 0.4 59.7 43.1 55.3 53.3 3.7 31.8 10.1 5.8
Median 9.1 43.9 1.0 3.9 62.5 0.3 58.4 33.7 59.6 53.4 3.7 9.6 3.8 2.8
Min 0.3 2.3 d.l. 0.1 0.1 d.l. 52.3 0.5 0.5 51.7 0.3 1.2 0.7 0.6
Max 145.8 467.3 2.5 47.9 404.3 1.5 71.7 182.9 199.8 54.6 11.0 118.6 35.5 22.9
1st quartile 1.1 15.6 0.6 1.0 15.6 0.2 53.1 14.2 9.5 53.4 0.8 6.1 2.2 1.0
3rd quartile 29.6 83.0 1.5 13.3 206.1 0.6 62.5 49.9 79.7 53.4 4.3 41.1 10.0 6.1

Py II Mean 80.6 329.7 3.0 9.4 3,015.6 1.1 48.6 138.3 511.8 52.3 10.7 121.0 218.3 91.7
Median 3.7 117.6 2.4 5.9 148.0 0.6 49.0 40.3 69.9 52.6 3.7 25.9 31.3 6.8
Min d.l. 0.8 d.l. d.l. d.l. d.l. 30.9 0.1 0.1 48.3 0.1 0.7 0.1 0.5
Max 1,273.2 1,636.6 13.6 43.2 21,417.7 4.4 76.1 800.2 7,205.3 55.2 169.6 640.1 1,430.9
1,936.3
1st quartile 1.4 16.1 0.9 3.0 48.4 0.3 40.7 11.8 2.7 52.1 2.2 12.8 11.0 1.1
3rd quartile 19.0 417.2 3.2 13.9 533.6 1.6 54.5 124.2 557.2 53.6 8.7 102.1 142.7 40.4

Enargite
Mean 101.6 13.5 0.5 33.8 0.0 32.7 6,410.8 3.1 141.0
Median 108.5 13.4 0.6 33.2 0.0 32.7 5,527.7 3.2 135.9
Min 31.1 11.8 0.1 24.9 0.0 32.7 4,273.7 1.8 79.5
Max 214.5 15.5 0.8 45.8 0.1 32.7 9,543.1 5.5 211.3
1st quartile 52.7 12.5 0.4 28.8 0.0 32.7 4,876.6 2.6 95.1
3rd quartile 114.9 14.2 0.7 37.2 0.1 32.7 8,167.5 3.4 188.9

Note: See digital Appendices 1 to 4 for raw data

 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1721

10-2 10-2
Pyrite I
Pyrite II

As (afu)
Aa (afu)
10-4 10-4

10-6 10-6
10-1 10-1

Cu (afu)
Cu (afu)

10-3 10-3

10-5 10-5

10-3 10-3

Te (afu)
Te (afu)

10-5 10-5

10-7 10-7
10-3 10-5

Sb (afu)
Se (afu)

10-5
10-5

10-7

10-7
10-1
10-4
10-3
Ag (afu)

10
Co (afu)
-5

10-5
10-6

10-7 10-7

10-9 10-8 10-7 10-6 10-8 10-7 10-6 10-5 10-4

Au (afu) Ag (afu)
Fig. 11. Binary plots of trace element concentration in pyrite analyzed by LA-ICP-MS and shown in atoms per formula
unit (afu).

and Se covary and do so independently of Au, As, and Te, and
that Ag and Pb covary independently of the other two groups
of elements. It should be noted, however, that these group-
ings are not observed in all crystals. For example, Te behaves
independently of Au and As in some crystals and covaries with
Se in others. Nonetheless, the Cu group of Cu, Ni, and Co
generally covaries independently of the Au group of Au, As,
and Te (Se). Silver concentration is usually decoupled from
those of both Au and Cu, and commonly covaries with Pb, Bi
and, rarely, Se concentration.
Enargite: On the basis of quantitative EMP analyses
(Table A3), enargite has close to an end member composi-
tion with a median As concentration of 18.5 wt % (IQR
= 18.118.8 wt %), a median S concentration of 32.8 wt %
(IQR = 32.733.0 wt %) and a median Cu concentration of
48.6wt% (IQR = 48.148.8wt %). The trace element con-
centrations were analyzed by a combination of EMP and LA-
ICP-MS methods, and predictably, Sb was the trace element
with the highest concentration with a median of 5527 ppm
(IQR = 48768167 ppm; Tables 3, A4). This is well below its
concentration in stibioenargite (Springer, 1969; Maske and
Skinner, 1971; Posfai and Buseck, 1998). Iron is the next
most important trace element with a median concentration of
316 ppm (IQR = 117649 ppm), followed by Te and Se, with
median concentrations of 135 ppm (IQR = 95.1189 ppm)
and 3.2ppm (IQR = 2.63.4 ppm), respectively (Table 3). The
1722 KING ET AL.

BSE Images Cu map

A B

Composite element maps

C D

Arsenic
Copper
Iron
E F

Arsenic
Copper
Iron

G H
Selenium
Copper
Iron
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1723

Au content is significantly lower than that of py I or py II, with
a median of 0.6 ppm (IQR = 0.40.7 ppm), but the median
silver content is significantly higher than that of either gen-
eration of Py (108 ppm, IQR = 52.7114.9 ppm). From BSE
images, the enargite appears to be unzoned (Fig. 8C), but LA-
ICP-MS depth profiles show small variations in the signals of
some trace elements and Cu (Fig. 13). Despite the presence
of a number of trace elements in significant concentration in
the enargite, none of them display a systematic correlation
with another trace element.
Table 4. Sulfur Isotope Compositions of Py I and Py II
from the Bawone and Binebase Deposits
Samples 33S 34S 36S
BID16-66.3 PyI 2.94 5.73 10.82
BOD3-81.5 PyII 2.54 4.93 9.59
BOD1-99.4 PyI 2.48 4.86 9.39
Note: The full analytical uncertainty (1s) for sulfur isotope analysis is
estimated to be 0.13 for 34S values, based on the long-term standard
deviation for repeat analyses of in-house standards

Sulfur Isotopes
Isotopic compositions of pyrite and sulfate (alunite and bar-
ite) were measured to better understand the physiocochemi-
cal conditions of ore formation. In particular, they were used
to estimate temperature and assist in the interpretation of fO2
and pH conditions.
Method
Because it is impossible to physically separate the fine-
grained, intergrown pyrite and alunite, separation was
achieved chemically in two stages. In the first stage, finely
ground altered rock was reacted with a Cr-reducing solu-
tion to liberate H2S from the sulfides (Canfield et al., 1986).
This procedure stripped the sample of sulfide, allowing the
solid residue to be filtered and reacted with Thodes solution
(Thode et al., 1961) to convert the remaining sulfur (sulfate)
to H2S. In both cases, the resulting H2S was trapped using a
zinc-acetate solution and reacted with a solution of AgNO3
to produce solid Ag2S that was purified and reacted with SF6
(Thode and Rees, 1979) prior to introduction into a Thermo
Finnigan MAT 253 dual-inlet gas-source mass spectrometer
at McGill University for measurement of sulfur isotope ratios,
with IAEA-S-1 as an internal standard (Sharman, 2011). The
ratios are reported relative to V-CDT in notation.
Results
Two samples of argillically and advanced argillically altered
material and one sample of pyrite and barite from the late
barite-enargite veins were selected for sulfur isotope analysis
(Table 4). The sulfate concentration in the preliminary analy-
sis was below the detection limit but consistent 34S values
were obtained for pyrite. The 33S, 34S, and 36S per mil val-
ues for pyrite average 2.65 ( 1s = 0.25), 5.17 ( 1s = 0.48),
and 9.93 ( 1s = 0.77), respectively. The standard deviations
are small considering that the pyrite represents different gen-
erations of mineralization.
Discussion
Geologic setting
The Bawone and Binebase deposits occur at the southeast-
ern end of a young volcanic island with an active volcano to
the northwest and a chain of extinct/dormant volcanic edifices
to the south. Both deposits are hosted by rocks of the Bine-
base Group, a thick andesitic sequence of alternating, inter-
bedded porphyritic lava flows or epiclastic deposits and crystal
tuffs that were probably deposited in close proximity to a vol-
canic center. If, as seems likely, a volcano was located above
the Binebase and Bawone deposits (and was thus close to the
coast), its erosion would have been very rapid due to a combi-
nation of rising sea level (as indicated by the marine deposits
of the overlying Pinterang Formation), thermal subsidence as
the magmatic center moved northwest, and the hot, wet trop-
ical climate. Consequently, all traces of the volcanic edifice
could have disappeared in tens of thousands of years.
Paleodepth and pressure
A very crude, approximate estimate of the depth and pres-
sure of emplacement of the Bawone and Binebase deposits
can be made by comparing their current elevation to that of
the base of the crater of Awu volcano, the only active volcano
on Sangihe Island. This assumes that the volcano overlying
these deposits reached an elevation similar to that of Awu vol-
cano, and although this cannot be known, it is suggested by the
crater elevation of the somewhat larger dormant/extinct Kaki-
raeng volcano located immediately to the west of the deposits.
The elevation of the crater of this volcano is 1,006 m (it has
likely been reduced in height due to erosion), whereas that
of the crater of Awu volcano is 1,327 m (Fig. 1). By contrast,
the Bawone and Binebase deposits are exposed at elevations
of 86 and 65 m, respectively. Thus, if the crater of the volcano
above the deposits was at an elevation similar to those of Awu

Fig. 12. BSE images (A, C, E, G) and corresponding copper map (B) or composite Cu (green), Fe (blue) and As/Se (red)
maps (D, F, H) showing strong sector and growth zoning in py II. Brighter colors correspond to higher concentrations of the
given element. (A) BSE image of complex, m-scale, strongly growth zoned drusy pyrite overgrowing massive, porous pyrite,
also zoned. (B) Cu map of the same crystal, light blue-green represents higher Cu concentration; the darker the blue, the
lower is the Cu concentration. The growth zoning is perpendicular to the crystal growth direction and records the primary
growth history (50 m scale). (C, D) Drusy, py II with growth and sector zones and very high Cu concentrations (green)
and variably As-rich (red) bands transitioning outward into massive, porous py II with lower Cu concentration and isolated
zones of As enrichment (scale bar is 50 m). (E, F) Cross section is through a euhedral py II crystal displaying growth zoning
with alternating As (red) and Cu (green) enriched bands. Sector zoning most strongly evident as a Cu enrichment on three
equivalent faces about a three-fold axis (scale bar is 500 m). (G, H) Drusy py II crystals displaying complex growth zoning in
either Cu (green) or Se (red) bands. Sector zoning in Cu is also present in the form of a cross at the center of a cubic pyrite
grain, top-center (scale bar is 500 m).
1724 KING ET AL.
Growth Zone 1
Co59
1 000 000
Fe57
Cu65
100 000 S34
10 000 Ni60
Signal Intensity (counts)
Se77
1000
Bi209
Sb121
100 Zn66
Ag107
10 Pb208
50
Fig. 13. An example of a depth profile through three growth zones of a py II crystal produced by LA-ICP-MS, in counts vs.
time (depth) for major and trace elements. Arsenic, Te, and Au covary through the three growth zones and are independent
of Co, Cu, Ni, Zn, Se, and Bi, whereas Ag, Pb, and to some extent Sb covary independently of the other two groups.
and Kakiraeng volcanoes, it follows that the deposits formed
at depths of approximately 900 to 1,300 m below the paleo-
vent. These depths fall within the range of depths estimated
for other high-sulfidation epithermal deposits (Jannas et al.,
1990; Arribas, 1995; Cooke and Simmons, 2000). The depths
inferred for Bawone and Binebase (9001,300m) correspond
to lithostatic pressures of 250 to 350 bars or hydrostatic pres-
sures of 90 to 125 bars (assuming the water table was near the
paleosurface), respectively. As much of the mineralization is
disseminated and there is no evidence of large through-going
fractures that might have intersected the paleosurface, pres-
sure was probably intermediate between the two extremes. In
the discussion that follows, we therefore assume that the pres-
sure accompanying hydrothermal alteration and gold mineral-
ization was ~200 bars.
Temperature and pH
Earlier, we showed that advanced argillic (AA) and inter-
mediate argillic (IA) alteration in the Bawone and Binebase
deposits are characterized by the following mineral associa-
tions: AA I (quartz + pyrite + pyrophyllite natroalunite), AA
II (quartz + pyrite natroalunite dickite kaolinite), AA III
(quartz + pyrite + kaolinite + alunite), IA I (quartz + kaolinite
pyrite dickite), and IA II (quartz + illite pyrite). Here, we
use stability relationships among these minerals in the system
Al-Si-O-K-Na-Fe-H-S to make inferences about the tempera-
ture and pH of the fluids responsible for the different types of
Growth Zone 2 Growth Zone 3
As75
Te125
Au197
100 150 200 250
Time (s)
alteration (Figs. 14, 15). The total S concentration was assumed
to be 0.01 m, consistent with that inferred for other high-sul-
fidation epithermal systems (e.g., Muntean et al., 1990). At a
pH below 2, aluminum is mobile, an observation confirmed by
our mass balance calculations that provides evidence for local
Al mobility. For these conditions, Al activity was assumed to be
0.1 (Knight, 1977; Fulignati et al., 1999). At higher pH, Al was
assumed to be immobile and its activity buffered by the alumi-
num silicate minerals (Stoffregen, 1987; Salvi et al., 1998).The
Na/K ratio was taken to be 10, assuming equilibrium between
natroalunite and alunite (natroalunite and alunite commonly
occur together but their textural relationships could not be
determined because of the very fine-grained nature of the
rock), and the conclusion that alunite transformed to natroal-
unite at a temperature of ~275C (see below). Oxygen fugac-
ity was inferred from the 34S values for pyrite to be that of
the hematite-magnetite buffer (see below). The activities of
Fe and Si were assumed to have been buffered by pyrite and
quartz, respectively, which are ubiquitous in the deposits.
At the conditions for which Figures 14 and 15 were con-
structed, pyrite is stable above 250C (hematite is stable
below 250C), making this the lower temperature limit for all
alteration facies. Because kaolinite and quartz react to form
pyrophyllite at temperatures greater than 280C (Fig. 15), the
coexistence of quartz and kaolinite in AA II, AA III, and IA
I alteration zones constrains their temperatures to have been
less than 280C. Conversely, the presence of pyrophyllite in
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1725

400

-
4

SO 2-
HSO

4
And

350
Py Mag
Na-Al Pyr

300
1 2 3
Musc Albite

4 5 Microcline
T (C)

250
Hem
Fe2+

Al Kao
Range of alteration
200 conditions
AA I fields 1 & 2
AA II field 1
AA III fields 4 & 5
150
IA I field 5

IA II field 3
100
0 2 4 6 8 10 12 14
pH
Fig. 14. Stability relationships among minerals in the K-Na-Al-Si-O-H and Fe-S-O systems as a function of temperature
and pH. Also shown are the fields of stability of the five main alteration facies (shaded areas), advanced argillic I (AA I),
advanced argillic II (AA II), advanced argillic III (AA III), intermediate argillic I (IA I), and intermediate argillic II (IA II).
The diagram was calculated for 200 bar (SVP), log fO2 = 30, quartz saturation, Na+/K+ = 10, aS = 0.01 and aAl3+ = 0.1 (see
text for details). Silicate and sulfate species are shown by solid lines, Fe species by dashed lines and sulfur species by dotted
lines. The model was calculated using SULF34 (Zhang, 1993) and The Geochemists Workbench (GW; Bethke, 2005) with
thermodynamic data from Holland and Powell (1998), Stoffregen et al. (2000), Helgeson (1978), Robie and Hemingway
(1995), and Johnson et al. (1992).

the AA I alteration zone indicates that this alteration occurred
at a temperature above 275C (Holland, 1998). Natroalunite
is the stable form of the alunite group minerals in the AA I
alteration zone, and alunite is the stable form in the AA III
alteration zone. Unexpectedly, however, given the coexistence
of quartz and kaolinite (implying a temperature less than
280C), natroalunite is also the stable member of the alunite
family in the AA II alteration zone. We interpret this to indi-
cate that AA II alteration occurred at ~280C, an interpreta-
tion that is supported by the presence of dickite in the AA II
alteration zone (kaolinite is widely inferred to convert to dick-
ite above 150C: Hemley et al., 1969; Stoffregen and Alpers,
1987; Stoffregen and Cygan, 1990). In view of the distribution
of alunite group minerals among the AAI, AAII, and AAIII
zones, we conclude that alunite was replaced by natroalunite
at ~275C.
An upper temperature limit for AA I alteration is provided
by the absence of andalusite, which forms from the reaction of
pyrophyllite and quartz at ~340C (Figs. 14, 15). In summary,
we conclude that AA I alteration occurred at a temperature
between 275 and 340C, AA II alteration at ~275C, and AA
III and IA I alteration at between about 250 and 275C. As
the IA II alteration association includes pyrite, this alteration
occurred at >250C and, although we cannot constrain the
upper temperature from its mineralogy, it is reasonable to
conclude that this limit was less than or similar to that for IA I
alteration, i.e., 275C.
On the basis of the stability relationships illustrated in Fig-
ure 14, specifically the presence of both natroalunite and
pyrophyllite in the AA I association, natroalunite and kaolin-
ite/dickite in the AA II association, and alunite and kaolinite in
the AA III association, we interpret the pH of advanced argil-
lic alteration to have been ~2.5. In principle, the pH could
have been significantly lower, if natroalunite and pyrophyllite
or alunite and kaolinite were not in equilibrium (Hemley and
Jones, 1964; Knight, 1977). If, however, the pH was below 2,
aluminum would have been highly mobile (Stoffregen, 1987),
which based on our analysis of mass changes (Fig. 10) was not
the case; aluminum experienced only minor depletion. This
supports a conclusion that the pH was at least 2 and more
likely 2.5. The presence of kaolinite in the IA I alteration
association constrains its pH to between 2.5 and 4, and the
presence of illite, but absence of K-feldspar and albite in the
IA II association, constrains its pH to between ~4 and 6.
1726 KING ET AL.

A -20
250C 15

HSO4-
SO42-
AlSO4+
-25 Al Kao Musc Microcline 10

log fO2 (HM)

log fO2
-30 5
Hem

34H
S= -1 Mag
2
0
-35 34H 0
2 S = -1
Py
34H
2 S= -5

H 2S
HS-
Fe 2+
Pr
-40
0 2 4 6 8
pH
B -20 10

HSO4-
AlSO4+ 300C

SO42-
Na-Al Pyr Musc Albite
-25

log fO2 (HM)

5
SO2

Hem
log fO2

-30 Mag 0
34H
S= -1
Py
2
0
34H
2 S= -1
34H
2 S= -5

-35 -5
H 2S
HS-

Fe+2 Pr
-40
0 2 4 6 8
pH
Fig. 15. Log fO2-pH diagrams showing stability relationships for minerals in the K-Na-Al-Si-O-H and Fe-S-O systems (A)
at 250C, and (B) at 300C. The diagrams were created assuming a pressure of 200 bar (SVP), quartz saturation, Na+/K+ = 10,
aS = 0.01 and aAl3+ = 0.1 (see text for details). Silicates and sulfate species are shown by solid lines, Fe-species by dashed lines
and sulfur species by dotted lines. The red lines are 34SH2S contours at 1, 5 (the measured 34S value for Sangihe pyrite)
and 10, and constrain the range of log fO2 values for Au-Ag mineralization to approximately 32 (Fig 16A) and 27 (Fig 16B)
assuming a pH of 2.5 (see text for details). The term Dlog fO2 (HM) refers to the number of log units of fO2 above or below that
of the hematite-magnetite buffer (calculated using The Geochemists Workbench).

fO2 Hedenquist et al., 1998b; Bethke et al., 2005; Deyell et al.,

Ore-forming high-sulfidation epithermal systems are char-
acteristically oxidizing, with oxygen fugacity straddling or just
below that of the hematite-magnetite buffer. For example, val-
ues for log fO2 (HM) reported for some other high-sulfidation
deposits have ranges of 4.24 to 1.32 (Thiersch et al., 1997),
2.4 to 0.07 (Muntean et al., 1990) and <2.5 (Voudouris, 2011).
If the sulfur source were dominantly magmatic, which is con-
sidered to be the case for most high-sulfidation deposits (e.g.,
Muntean et al., 1990; Rye et al., 1992; Rye, 1993; Arribas, 1995;
2005a, b; Fifarek and Rye, 2005; Taylor, 2007), the S34S value
of the fluid would have been approximately zero. Using this
assumption, it is possible to calculate the 34S of the fluid from
the measured pyrite 34S values and, in turn, calculate log fO2
for given pH values (Zhang, 1993). The negative 34S values
calculated for reduced sulfur (H2S) at Bawone and Binebase,
assuming a temperature between 250 and 340C, are simi-
lar to those of other high-sulfidation deposits (Muntean et
al., 1990; Hedenquist et al., 1994; Voudouris, 2010), and are
consistent with sulfur isotope fractionation between sulfides/
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1727
sulfosalts and sulfates in an oxidizing environment (Arribas,
1995; Cooke and Simmons, 2000). The log fO2 (HM) values
corresponding to the 34SH2S values for advanced and interme-
diate argillic alteration are between 4 and 1 (Fig. 15; Table 5;
Rye, 1993, 2005; Cooke and Simmons, 2000).
Metal transport
According to many researchers, there is a hyper-acidic
ground preparatory or pre-ore stage of hydrothermal alter-
ation in high-sulfidation epithermal systems that precedes gold
mineralization and results in advanced argillic alteration (Stof-
fregen, 1987; Hedenquist et al., 1994, 1998, 2000; Arribas,
1995). There is also a consensus among these researchers that
the extreme acidity required to produce the advanced argillic
alteration that characterizes this ground preparation stage can
only be explained by interaction of the rocks with a condensed
acidic vapor (e.g., Hedenquist et al., 1994, 2000). At Bawone
and Binebase, the alteration does not reflect interaction with
fluids having the extreme acidity of some high-sulfidation
epithermal systems (residual vuggy silica is not observed).
Nonetheless, the very low pH of ~2.5 estimated for the fluid
requires that it was a liquid condensed from gas, as other fluids
that might have been present in the system, such as meteoric
water/groundwater, seawater and magmatic liquid would have
had higher pH (Stoffregen and Alpers, 1987; Meyer and Hem-
ley, 1967; Hemley et al., 1969; Stoffregen, 1987; White, 1957).
Because advanced argillic alteration involved crystallization of
auriferous and argentiferous pyrite (pyrite I) and this pyrite
accounts for between 30 and 50% of the hypogene Au (Ag-Cu)
mineralization in the Bawone and Binebase deposits, it there-
fore follows that much of the gold and the other metals were
introduced into the deposits by this condensed gas. Whether
or not pyrite II and enargite, which were introduced later, also
crystallized from this condensed gas is less clear but given the
similar trace element and isotopic compositions of pyrite I and
II, it is a strong possibility. There were also likely multiple iter-
ations of this process as porosity was created and subsequently
filled by alteration products (Berger and Henley, 2011).
Evidence that metals can be transported in high concentra-
tions in magmatic vapors has been provided by analyses of the
compositions of vapor inclusions, which have yielded poten-
tially ore-forming concentrations of some metals, including
ppm levels of Au and Ag (Audtat et al., 1998, 2008; Ulrich et
al., 1999; Williams-Jones and Heinrich, 2005; Seo et al., 2009).
In addition to the fluid inclusion data, there is also a growing
body of evidence from experiments that Au, Ag and Cu can be
transported in appreciable concentrations in aqueous vapor
(Migdisov et al., 1999; Archibald et al., 2001, 2002; Migdisov
and Williams-Jones, 2013; Hurtig and Williams-Jones, 2014).
Table 5. Physicochemical Conditions for Each Alteration Facies Based
on Stability Relationships Among Minerals and 34S Values for Py I and Py
II
T pH log fO2 log fO2 (HM buffer)
IA I 260275 34 34 to 31 +1 to +3
IA II >265 46 35 to 33 1 to +2
AA I 275330 0.54 28 to 27 +2 to +4
AA II 270275 0.54 28 to 27 +1 to +4
AA III 250270 0.54 33 to 32 +1 to +3
For example, Migdisov and Williams-Jones (2013) and Hur-
tig and Williams-Jones (2014) have shown that HCl-bearing
aqueous gases can dissolve ppm levels of Ag and Au, respec-
tively, at temperatures as low as 400C. They have shown,
moreover, that the solubility of these metals increases expo-
nentially with increasing f H2O. At the temperature and pres-
sure predicted for the formation of the Bawone and Binbase
deposits, i.e., close to those of the H2O vapor-liquid boundary,
conditions would have been optimal for the transport of gold
and silver by magmatic vapor.
Direct evidence that condensed gases in environments
analogous to those of high-sulfidation epithermal deposits
can transport significant concentrations of metals is provided
by Kawah Ijen, an active volcano in eastern Java, Indonesia.
There, advanced argillic alteration (and residual silica altera-
tion) is ongoing and accompanied by the crystallization of
pyrite enriched in Cu, Ag, and Au (Scher et al., 2013). The
fluid responsible for this alteration is a condensed magmatic
gas with a pH <1 containing ppm concentrations of Cu and
As (the concentrations of Au and Ag are below the level of
detection). Although the concentrations of Cu, Ag, and Au
in the Kawah Ijen pyrite are much lower than those in the
pyrite at Bawone and Binebase (because of the much lower
pressure and hence density of the gas), the metal ratios, e.g.,
Cu/Au and Ag/Au are very similar to those of the two Sangihe
Island deposits (Fig. 16). We can therefore conclude that the
two pyrite occurrences likely formed by the same process and
that the fluid which transported the ore metals in the Bawone
and Binebase deposits was therefore a magmatic gas. Several
other ore deposit studies have also documented gold mineral-
ization associated with early advanced argillic alteration (e.g.,
Muntean et al., 1990; Voudouris, 2011).
Ore deposition
For the reasons presented above, we propose that the fluid
responsible for the formation of the Bawone and Binebase
deposits was a condensed magmatic gas. However, we con-
sider it likely that this gas encountered a physicochemical
barrier, such as the water table or an impermeable cap rock.
This seems to have been the case for the deeper mineraliza-
tion at Bawone (Figs. 3A, 6A), as the gas condensed to a low-
pH liquid from which the ore mineral (pyrite) precipitated
or formed by replacement of primary iron-bearing minerals.
According to this interpretation, the ore metals were trans-
ported in the vapor as hydrated species, e.g., MeCl-H2O or
MeHS-H2O (Migdisov et al., 1999; Williams-Jones and Hein-
rich, 2005; Zezin et al., 2011b; Migdisov and Williams-Jones,
2013; Hurtig and Williams-Jones, 2014), whereas after con-
densation of the vapor to liquid, they would have been trans-
ported as charged or neutral aqueous species like MeCl2 or
MeHSo (Crerar and Barnes, 1976; Gammons and Barnes,
1989; Zotov et al., 1990; Gammons and Williams-Jones, 1997;
Mountain and Seward, 2003; Stefnsson and Seward, 2004;
Williams-Jones et al., 2009).
If saturation of the fluid with the metal had occurred, ore
deposition would have been controlled by reactions such as
the following:
Au(HS)o + H2O(l) = Au(s) + H2S + O2 (1)
AuCl2 +
+ H2O(l) = Au(s) + 2Cl + H + O2(g) (2)
1728 KING ET AL.

Pyrite (Scher et al., 2013) is adsorbed at a pH <5 (Widler and Seward, 2002). Pyrite can
Pyrite I also have positively charged faces if arsenic bearing (Favorov
1000 Pyrite II et al., 1974), but the effects of Cu on the surface properties of
pyrite are unknown.
As (ppm)

100 Pyrite I is an integral part of advanced and intermediate

argillic alteration, and accounts for 30 to 50% of the gold in
10 the Bawone deposit and in the hypogene part of the Bine-
base deposit with an median concentration of 1.0 ppm Au;
1
the median concentrations of Ag and Cu are 9.1 ppm and
0.44wt%, respectively. The texturally complex pyrite II con-
tains median values of 2.4 ppm Au, 3.7 ppm Ag, and 1.1 wt %
Cu, and accounts for most of the remaining gold. Some gold
and even more silver are present in enargite (median values of
1000 0.6 ppm Au and 108 ppm Ag).
Insights into the conditions favorable for the uptake of gold
by pyrite in the Bawone and Binebase deposits are afforded
Cu (ppm)

100
by the presence of both growth and sector zoning in pyrite
II. The latter zoning reflects differences in the concentrations
10
of elements on two nonequivalent faces within single growth
zones, and enables partition coefficients to be calculated
1 for these faces, which record the relative changes in physi-
cochemical conditions during crystal growth (van Hinsberg
and Schumacher, 2007). We have used analyses of sector and
growth zones in pyrite II to determine the conditions favor-
1000 able for the incorporation of gold and other trace elements
into pyrite. The element concentrations in two adjacent
100
nonequivalent faces were measured for a series of growth
Ag (ppm)

10 zones and the partition coefficients (Kd) were calculated. We

1 periods of growth marked by relatively constant Kd, and lower
0.1
0.01
0.0001 0.001 0.01 0.1 1 10
Au (ppm)
Fig. 16. Log-log plots comparing the trace element composition of py I
and py II with that of pyrite from an active high-sulfidation epithermal system
at Kawah Ijen volcano (Scher et al., 2013). From these plots, it is evident that
the ratios, Cu/Au and Ag/Au of py I and py II, and particularly the Ag/Au ratio
of py I, are broadly similar to those of Kawah Ijen pyrite.
However, as native gold does not occur in the deposits,
instead, the gold occurs with other metals as nanoparticles in
pyrite or within the structure of this mineral, it follows that
gold was undersaturated in the liquid.
Deposition of gold and other metals is interpreted to have
occurred as a result of their adsorption onto the faces of pyrite
during crystal growth. This deposition began at the onset of
advanced argillic alteration with the precipitation of pyrite I
and continued with the precipitation of pyrite II. Adsorption
of gold onto pyrite has been proposed previously as a mecha-
nism to explain the occurrence of invisible gold in other epi-
thermal deposits (e.g., Simon et al., 1999; Wilder and Seward,
2002; Pals et al., 2003). It also has been investigated experi-
mentally at ambient pressure and temperature for pH values
from 2 to 10. Results of these experiments for pyrite having
negatively charged crystal faces show that over 90% of the
gold in solution (undersaturated with respect to native gold)
observed that the highest gold concentrations coincided with
gold values coincided with sharp increases and decreases in
the Kd (Fig. 17). Thus, in contrast to mineralization produced
by saturation of an ore mineral, which is favored by rapid
changes in the physicochemical conditions due to processes
like mixing and boiling, concentration of gold in the Bawone
and Binebase deposits was favored when physicochemical
conditions were relatively stable.
Incorporation of trace elements in pyrite
Unlike many examples of deposits of invisible gold, the
pyrite in the Bawone and Binebase deposits contains only
trace quantities of As (Tables 3, A1, A2). Despite this low
concentration of As relative to that in auriferous pyrite from
other deposits (e.g., Reich et al., 2005), Au concentration cor-
relates positively with the concentration of As (Figs. 11, 13,
16); the latter is a hundred times greater than the concentra-
tion of Au in py I and py II. This correlation suggests that
Au + As substituted for Fe, as has been proposed previously
for other deposits containing gold hosted in pyrite (Cook and
Chryssoulis, 1990; Fleet et al., 1993; Huston et al., 1995). It
is also possible, however, that some or all of the gold occurs
as nanoparticles within the pyrite. Although our data do not
allow us to determine whether the gold was incorporated into
the pyrite structure or is present as nanoparticles, Reich et al.
(2005) have observed that in pyrite with Au/As > 0.02, gold
occurs as microinclusions, whereas if the ratio is less than this,
it is substituted into the pyrite structure. The average Au/As
for pyrite in the Bawone and Binebase deposits is 0.019, sug-
gesting that much of the gold may have been incorporated in
the structure of this mineral.
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1729

16.79 Kd
Cu light
Au (ppm)

4
3

11.46 12.22
3

Cu (wt %)
2
Kd

1
1

2.65 2.23

0 0

0 20 40 60 80 100
Distance
Fig. 17. A plot showing the variation in the partition coefficient (Kd; black line) and the concentration of Cu (red line)
as a function of the distance (in percent) from the core to the rim of a single pyrite crystal (BID34-84). In this plot, the Kd
corresponds to the ratio of the Cu concentration in one sector zone (the one with the higher overall Cu concentration, i.e.,
the light zone in BSE images, e.g., Fig. 13) to the Cu concentration in the adjacent sector zone (the dark zone), within
the same growth zone. The Kd is inferred to vary inversely with temperature following van Hinsberg and Schumacher (2007).
The yellow squares represent gold concentrations analyzed by LA-ICP-MS, with the width of the square corresponding to
the diameter of the pit in relation to the growth zones. The vertical scale on the right hand side of the diagram indicates the
Cu concentration. The highest gold concentrations correspond to regions where the Kd is stable, whereas lower gold concen-
trations correspond to intervals over which there is a rapid change in Kd values; Cu concentrations show the same behavior.

Incorporation of Au in the pyrite structure is also favored
by the presence of Te, which substitutes for the smaller S
ion, thereby expanding the lattice and providing space for the
large Au ion (Chouinard et al., 2005a; Bi et al., 2011). The
presence of appreciable Te (concentrations of Te are tens (py
I) to a hundred (py II) times greater than that of Au) and its
positive correlation with Au (Fig. 14) suggests that this substi-
tution was also important in the uptake of Au.
Although the As concentration in pyrite II is low compared
to that of auriferous pyrite elsewhere, the Cu concentration is
anomalously high, up to 6 wt % Cu (Abraitis et al., 2004). Sig-
nificantly, most of the other examples of auriferous pyrite with
high copper contents are also from high-sulfidation epithermal
deposits, e.g., Pascua Lama (up to 1.5 wt % Cu; Chouinard
et al., 2005a), Pueblo Viejo (up to 3.0 wt % Cu; Deditius et
al., 2011) and Chelopech (up to 4.5 wt % Cu; Pacevski et al.,
2008). Moreover, the exceptions, such as pyrite from the Coka
Marin polymetallic deposit (up to 8 wt % Cu; Pacevski et al.,
2008) have many characteristics of high-sulfidation deposits.
It has been shown that copper can be incorporated in
the pyrite structure both as nanoparticles and by substitu-
tion for Fe, even within a single ore deposit such as Peublo
Viejo (Huston et al., 1995; Oberthr et al., 1997; Deditius et
al., 2011). However, there is no agreed upon mechanism to
explain the rare and anomalous Cu concentrations in pyrite.
Some studies have shown that Cu-rich pyrite is the result of
recrystallization of Cu-phases (Pacevski et al., 2008). This was
not the case at Bawone and Binebase because of the well-
documented zoning that records the growth history of the
crystals (Fig. 12). Other researchers have suggested oxidation
of aqueous Cu+ and direct substitution of Cu2+ for Fe2+ (e.g.,
Chouinard et al., 2005a; Pacevski et al., 2008) accompanied by
the incorporation of other trace elements that cause distortion
of the lattice (e.g., Radcliff and McSween, 1970). Whatever
the mechanism, it seems likely that the highly oxidizing condi-
tions that are characteristic of high-sulfidation deposits play a
role in the incorporation of Cu in the pyrite structure.
Genetic model
We propose that the metals forming the Bawone and Bine-
base deposits originated from a relatively oxidized magma that
was emplaced at high crustal levels and exsolved a low den-
sity supercritical fluid (that subsequently evolved to vapor)
into which metals (including Au, Ag, and Cu) partitioned
1730 KING ET AL.

preferentially. The gold and other metals were transported
as hydrated species that ascended through the volcanic pile
via fractures and zones of enhanced permeability to depths
between 900 and 1,300 m, where the vapor condensed at tem-
peratures between 250 and 340C to form an acidic liquid
with a pH of ~2.5; fO2 ranged up to four log units above the
hematite-magnetite buffer.
The acidic condensate altered the host andesite, replac-
ing primary mafic minerals, such as hornblende, biotite, and
magnetite, with auriferous py I (sulfidation), and converting
feldspars to an advanced argillic alteration association that
includes kaolinite, pyrophyllite, alunite, natroalunite, and
dickite. If fluids had been previously neutralized by reactions
with rocks or fluid/rock ratios were lower, the alteration led
to an intermediate argillic I alteration association of kaolin-
ite, quartz, and pyrite, and a distal intermediate argillic II
alteration association of illite, quartz, and pyrite. The altera-
tion also created porosity through the dissolution of minerals.
During alteration, all major elements, except for Zr and Ti
were depleted proximal to the main alteration/mineralization
conduits (Al and particularly Si experienced only minor deple-
tion; Fig. 10). Iron and many trace elements, except for the
REE, Rb and Cs, were enriched, in some cases by orders of
magnitude, relative to their original abundance.
Auriferous py II was deposited in veins and in open space
created by mineral dissolution. These same veins were frac-
tured and the fractures filled by barite and enargite. Rare, late
auriferous (electrum) barite-base metal (sphalerite and chal-
copyrite) veins formed at Binebase.
Py I, which is associated with advanced and intermediate
argillic alteration, and py II, precipitated under conditions
for which metals, including gold, were undersaturated in
the fluid. The metals, i.e., Au, Ag, and Cu, were adsorbed
onto the surfaces of the growing pyrite crystals and incorpo-
rated in the pyrite either as a solid solution or as nanopar-
ticles. Some gold and silver were incorporated in enargite by
the same mechanisms (Fig. 18). The Bawone and Binbase
deposits were subsequently exhumed and partially oxidized,
creating an upper zone of supergene-enriched oxide ore and
deeper lower-grade hypogene ores.

AuHS . nH2O
Impermeable
H2 SO42-
AuHSo Impermeable H2 SO42-
AuCl2- H 2O
AuCl2-
Cu Cu
AuHS . nH2O
CuHS . nH2O Advanced Argillic
Cu
Pyrite
AuCl2 nH2O
.
Au
Py II Ba-En veins
AgCl . nH2O
H 2S SO2 AuHSo
CO2 H 2O CuHS. nH2O
Argillic

A B C

Fig. 18. A cartoon cross section through the stratovolcano that may have hosted the Bawone and Binebase gold deposits.
Moderately dipping flows and volcaniclastic units with variable porosity and permeability are crosscut by structures that
permitted transfer of heat and fluids (red) upward. The latter caused hydrothermal alteration of the adjacent rocks (pink).
Some of the conduits vented at the surface, producing solfataras, whereas others terminated at impermeable barriers, such as
the welded andesite tuff in the Bawone deposit, or at perched water tables corresponding to permeable and less permeable/
impermeable layers. The small circle in the cross section indicates the location of the large circles, labeled A to C. Circle A
illustrates the transport of the metals as hydrated species in magmatic vapor rising through the volcanic pile and condensing at
an impermeable boundary. Circle B shows the distribution of rocks altered by highly acidic condensed vapor, and the accom-
panying pyrite-hosted gold (copper, silver) mineralization. Finally, Circle C depicts a pyrite crystal with two types of faces
onto which copper and gold adsorbed preferentially from a gold-undersaturated condensed vapor to produce gold (copper,
silver)-rich sector- and growth-zoned pyrite II. Multiple iterations of this process are likely required to produce an economic
high-sulfidation deposit of this style.
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1731
Conclusions
Textural and geochemical observations provide compel-
ling evidence that the Bawone and Binebase gold deposits
were the products of a condensed acidic magmatic vapor.
This fluid produced the advanced argillic and intermediate
argillic alteration that is characteristic of high-sulfidation epi-
thermal systems and deposited gold, as well as copper and
silver, either within the structure of pyrite or as nanoparticles
within this mineral. The fluid did not deposit native gold or
electrum (except in very late veins at Binebase) and so we
conclude that it was undersaturated in respect to these met-
als. Instead of precipitating directly from the fluid, gold and
silver were concentrated by adsorption onto the surfaces of
growing pyrite (and to a much lesser extent enargite) crystals,
where they either formed nanoparticles or substituted for iron
in the structure of the mineral. Much of the auriferous pyrite
formed during advanced and intermediate argillic alteration
as a result of the sulfidation of primary mafic minerals. Aurif-
erous pyrite (py II) crystallized later, filling fractures in the
earlier formed pyrite, and was joined even later by Au-Ag
bearing enargite that precipitated in reopened pyrite II veins.
Although the later pyrite II is richer in Au than pyrite I, and
enargite is richer in Ag than both pyrite generations, the pre-
cious metals in both minerals are interpreted to have con-
centrated from a condensed magmatic vapor by adsorption
during crystal growth. The Bawone and Binebase deposits are
part of the continuum of high-sulfidation deposits, formed
from a condensed magmatic vapor, and record a mineralizing
process that likely has contributed to the formation of many
other high-sulfidation deposits.
Acknowledgments
The authors thank East Asia Minerals for their financial and
logistical support, in particular Tom Mulja for suggesting the
project and for his assistance and comments. The field com-
ponent of the research would not have been possible without
the help of Arodji Wisanggono, Johnnedy Situmorang, Mardy
Posumah, Grace Kapal, and the staff on Sangihe Island and
in Jakarta. Simon Jackson and Jeanne Percival at the GSC,
Boswell Wing and Libby Sharman of the Sulfur Isotope labo-
ratory at McGill University, Lang Shi and Jeanne Paquette
provided invaluable assistance with the analytical aspects of
the research. The project was funded by East Asia Minerals
and an NSERC-CRD grant to A.E. Williams-Jones and G.
Williams-Jones with additional support by DIVEX. Construc-
tive reviews by R. Fifarek, R.W. Henley, and associate editor
D. John improved the paper considerably.
REFERENCES
Abraitis, P.K., Pattrick, R.A.D., and Vaughan, D.J., 2004, Variations in the
compositional, textural and electrical properties of natural pyrite: A review:
International Journal of Mineral Processing, v. 74, p. 4159.
Agrawal, S., Guevara, M., and Verma, S., 2008, Tectonic discrimination of
basic and ultrabasic volcanic rocks through log-transformed ratios of immo-
bile trace elements: International Geology Review, v. 50, p. 10571079.
Archibald, S., Migdisov, A.A., and Williams-Jones, A.E., 2001, The stability of
Au-chloride complexes in water vapor at elevated temperatures and pres-
sures: Geochimica et Cosmochimica Acta, v. 65, p. 44134423.
Archibald, S.M., Migdisov, A.A., and Williams-Jones, A.E., 2002, An experi-
mental study of the stability of copper chloride complexes in water vapor at
elevated temperatures and pressures: Geochimica et Cosmochimica Acta,
v. 66, p. 16111619.
Arribas, A.R., 1995, Characteristics of high-sulphidation epithermal deposits,
and their relation to magmatic fluid, in Thompson, J.F.H. ed., Magmas,
fluids, and ore deposits, Mineralogical Association of Canada, Short Course,
v. 23, p. 419454.
Audtat, A., Gnther, D., and Heinrich, C.A., 1998, Formation of a mag-
matic-hydrothermal ore deposit: Insights with LA-ICP-MS analysis of fluid
inclusions: Science, v. 279, p. 20912094.
Audtat, A., Pettke, T., Heinrich, C.A., and Bodnar, R.J., 2008, Special Paper:
The composition of magmatic-hydrothermal fluids in barren and mineral-
ized intrusions: Economic Geology, v. 103, p. 877908.
Beaulieu, S., 2010, InterRidge global database of active submarine hydro-
thermal vent fields: prepared for InterRidge, Version 2.0: World Wide Web
electronic publication, http://www.interridge.org.IRvents.
Berger, B.R., and Henley, R.W., 2011, Magmatic-vapor expansion and the for-
mation of high-sulfidation gold deposits: Structural controls on hydrother-
mal alteration and ore mineralization: Ore Geology Reviews, v. 39, p. 7590.
Bethke, C.M., 2005, The Geochemists Workbench Release 6.0: GWB Essen-
tials Guide: University of Illinois: Urbana, Illinois.
Bethke, P.M., Rye, R.O., Stoffregen, R.E., and Vikre, P.G., 2005, Evolution
of the magmatic-hydrothermal acid-sulfate system at Summitville, Colo-
rado: Integration of geological, stable-isotope, and fluid-inclusion evidence:
Chemical Geology, v. 215, p. 281315.
Bi, S.-J., Li, J.-W., Zhou, M.-F., and Li, Z.-K., 2011, Gold distribution in As-
deficient pyrite and telluride mineralogy of the Yangzhaiyu gold deposit,
Xiaoqinling district, southern North China craton: Mineralium Deposita,
v. 46, p. 925941.
Canfield, D.E., Raiswell, R., Westrich, J.T., Reaves, C.M., and Berner, R.A.,
1986, The use of chromium reduction in the analysis of reduced inorganic
sulfur in sediments and shales: Chemical Geology, v. 54, p. 149155.
Chouinard, A., Paquette, J., and Williams-Jones, A.E., 2005a, Crystallo-
graphic controls on trace-element incorporation in auriferous pyrite from
the Pascua epithermal high-sulphidation deposit, Chile-Argentina: The
Canadian Mineralogist, v. 43, p. 951963.
Chouinard, A., Williams-Jones, A.E., Leonardson, R.W., Hodgson, C.J., Silva,
P., Tellez, C., Vega, J., and Rojas, F., 2005b, Geology and genesis of the
multistage high-sulfidation epithermal Pascua Au-Ag-Cu deposit, Chile and
Argentina: Economic Geology, v. 100, p. 463490.
Cook, N.J., and Chryssoulis, S.L., 1990, Concentrations of invisible gold in
the common sulfides: The Canadian Mineralogist, v. 28, p. 116.
Cooke, D.R., and Simmons, S.F., 2000, Characteristics and genesis of epith-
ermal gold deposits: Reviews in Economic Geology, v. 13, p. 221244.
Crerar, D.A., and Barnes, H.L., 1976, Ore solution chemistry; V, Solubilities
of chalcopyrite and chalcocite assemblages in hydrothermal solution at 200
to 350C: Economic Geology, v. 71, p. 772794.
Deditius, A.P., Utsunomiya, S., Renock, D., Ewing, R., Ramana, C., Becker,
U., and Kesler, S.E., 2008, A proposed new type of arsenian pyrite: Compo-
sition, nanostructure and geological significance: Geochimica et Cosmochi-
mica Acta, v. 72, p. 29192933.
Deditius, A.P., Utsunomiya, S., Ewing, R.C., Chryssoulis, S.L., Venter, D.,
and Kesler, S.E., 2009, Decoupled geochemical behavior of As and Cu in
hydrothermal systems: Geology, v. 37, p. 707710.
Deditius, A.P., Utsunomiya, S., Reich, M., Kesler, S.E., Ewing, R.C., Hough,
R., and Walshe, J., 2011, Trace metal nanoparticles in pyrite: Ore Geology
Reviews, v. 42, p. 3246.
de Vos, J., van den Hoek Ostende, L.W., and van den Bergh, G.D., 2007,
Pattern in insular evolution of mammals: A key to island palaeogeography,
in Renema, W. ed., Biogeography, Time, and Place: Distributions, barriers,
and islands: Dordrecht, The Netherlands, Springer, p. 315345.
Deyell, C.L., 2005a, Alunite in the Pascua-Lama high-sulfidation deposit:
Constraints on alteration and ore deposition using stable isotope geochem-
istry: Economic Geology, v. 100, p. 131148.
Deyell, C.L., Rye, R.O., Landis, G.P., and Bissig, T., 2005b, Alunite and the
role of magmatic fluids in the Tambo high-sulfidation deposit, El Indio
Pascua belt, Chile: Chemical Geology, v. 215, p. 185218.
Favorov, V.A., Krasnikov, V.I., and Sychugov, V.S., 1974, Variations in semi-
conductor properties of pyrite and arsenopyrite and their determinants:
International Geology Review, v. 16, p. 385394.
Fifarek, R.H., and Rye, R.O., 2005, Stable-isotope geochemistry of the Pie-
rina high-sulfidation AuAg deposit, Peru: Influence of hydrodynamics on
SO42H2S sulfur isotopic exchange in magmatic-steam and steam-heated
environments: Chemical Geology, v. 215, p. 253279.
Fleet, M.E., Chryssoulis, S.L., MacLean, P.J., Davidson, R., and Weisener,
C.G., 1993, Arsenian pyrite from gold deposits: Au and As distribution
1732 KING ET AL.
investigated by SIMS and EMP, and color staining and surface oxidation by
XPS and LIMS: The Canadian Mineralogist, v. 31, p. 117.
Fulignati, P., Gioncada, A., and Sbrana, A., 1998, Geologic model of the mag-
matic-hydrothermal system of Vulcano (Aeolian Islands, Italy): Mineralogy
and Petrology, v. 62, p. 195222.
Fulignati, P., Gioncada, A., and Sbrana, A., 1999, Rare-earth element (REE)
behaviour in the alteration facies of the active magmatic-hydrothermal sys-
tem of Vulcano (Aeolian Islands Italy): Journal of Volcanology and Geother-
mal Research, v. 88, p. 325342.
Gammons, C.H., and Barnes, H.L., 1989, The solubility of Ag2S in near-neu-
tral aqueous sulfide solutions at 25 to 300C: Geochimica et Cosmochimica
Acta, v. 53, p. 279290.
Gammons, C.H., and Williams-Jones, A.E., 1997, Chemical mobility of gold in
the porphyry-epithermal environment: Economic Geology, v. 92, p. 4559.
Garwin, S.L., 1990, South Sangihe Regional Reconnaissance Program:
Jakarta, Indonesia, Report for P.T. Meares Soputan Mining, 84 p.
Grant, A., 1986, The isocon diagram: A simple solution to Gresens equation
for metasomatic alteration: Economic Geology, v. 81, p. 19761982.
Hall, R., 1996, Reconstructing Cenozoic Southeast Asia, in Hall, R., and
Blundell, D. eds., Tectonic evolution of Southeast Asia: Geological Society,
London, Special Publication 106, p. 153184.
2000, Neogene history of collision in the Halmahera region, Indonesia,
in Proceedings of the Indonesian Petroleum Association 27th Annual Con-
vention, Jakarta, Indonesia, p. 487493.
Hedenquist, J.W., Matsuhisa, Y., Izawa, E., White, N.C., Giggenbach, W.F.,
and Aoki, M., 1994, Geology, geochemistry, and origin of high sulfidation
Cu-Au mineralization in the Nansatsu district, Japan: Economic Geol-
ogy, v. 89, p. l30.
Hedenquist, J.W., Arribas, A.R., and Reynolds, J.T., 1998, Evolution of an
intrusion-centered hydrothermal system: Far Southeast-Lepanto porphyry
and epithermal Cu-Au deposits, Philippines: Economic Geology, v. 93,
p. 373404.
Hedenquist, J.W., Arribas, A.R., and Gonzalez-Urien, E., 2000, Explora-
tion for epithermal ore deposits: Reviews in Economic Geology, v. 13, p.
245277.
Heinrich, C.A., Gnther, D., Audtat, A., Ulrich, T., and Frischknecht, R.,
1999, Metal fractionation between magmatic brine and vapor, determined
by microanalysis of fluid inclusions: Geology, v. 27, p. 755758.
Helgeson, H.C., 1978, Summary and critique of the thermodynamic proper-
ties of rock-forming minerals: American Journal of Science, v. 278, p. 1229.
Hemley, J.J., Hostetler, P.B., Gude, A.J., and Mountjoy, W.T., 1969, Some
stability relations of alunite: Economic Geology, v. 64, p. 599612.
Hemley, J.J., and Jones, W.R., 1964, Chemical aspects of hydrothermal altera-
tion with emphasis on hydrogen metasomatism: Economic Geology, v.
59, p. 538569.
Henley, R.W., and Berger, B.R., 2011, Magmatic-vapor expansion and the
formation of high-sulfidation gold deposits: Chemical controls on alteration
and mineralization: Ore Geology Reviews, v. 39, p. 6374.
Holland, T.J.B., and Powell, R., 1998, An internally consistent thermody-
namic data set for phases of petrological interest: Journal of Metamorphic
Geology, v. 16, p. 309343.
Hopf, S., 1993, Behaviour of rare earth elements in geothermal systems of
New Zealand: Journal of Geochemical Exploration, v. 47, p. 333357.
Hurtig, N., and Williams-Jones, A.E., 2014, An experimental study of the
transport of gold through hydration of AuCl in aqueous vapour and vapour-
like fluids: Geochimica et Cosmochimica Acta, v. 127, p. 305325.
Huston, D.L., Sie, S.H., Suter, G.F., Cooke, D.R., and Both, R.A., 1995,
Sulfide deposits: Part I. Proton microprobe analyses of pyrite, and part II.
Selenium levels in pyrite: Comparison with d34S values and implications
for the source of sulfur in volcanogenic hydrothermal systems: Economic
Geology, v. 90, p. 11671196.
Jannas, R.R., Beane, R.E., and Ahler, B.A., 1990, Gold and copper mineral-
ization at the El Indio deposit, Chile: Journal of Geochemical Exploration,
v. 36, p. 233266.
Johnson, J.W., Oelkers, E.H., and Helgeson, H.C., 1992, SUPCRT92: A soft-
ware package for calculating the standard molal thermodynamic properties
of minerals, gases, aqueous species, and reactions from 1 to 5000 bar and 0
to 1000C: Computers & Geosciences, v. 18, p. 899947.
Karakaya, N., 2009, REE and HFS element behaviour in the alteration facies
of the Erenler Dag Volcanics (Konya, Turkey) and kaolinite occurrence:
Journal of Geochemical Exploration, v. 101, p. 185208.
Kesler, S.E., Russell, N., Seaward, M., Rivera, J., McCurdy, K., Cumming,
G.L., and Sutter, J.F., 1981, Geology and geochemistry of sulfide mineral-
ization underlying the Pueblo Viejo gold-silver oxide deposit, Dominican
Republic: Economic Geology, v. 76, p. 10961117.
Knight, J.E., 1977, A thermochemical study of alunite, enargite, luzonite, and
tennantite deposits: Economic Geology, v. 72, p. 13211336.
Landtwing, M.R., Furrer, C.F., Redmond, P.B., Pettke, T., Guillong, M., and
Heinrich, C.A., 2010, The Bingham Canyon porphyry Cu-Mo-Au deposit:
III. Zoned copper-gold ore deposition by magmatic vapor expansion: Eco-
nomic Geology, v. 105, p. 91118.
Lewis, A.J., Palmer, M.R., Sturchio, N.C., and Kemp, A.J., 1997, The rare
earth element geochemistry of acid-sulphate and acid-sulphate-chloride
geothermal systems from Yellowstone National Park, Wyoming, USA: Geo-
chimica et Cosmochimica Acta, v. 61, p. 695706.
Lottermoser, B.G., 1992, Rare earth elements and hydrothermal ore forma-
tion processes: Ore Geology Reviews, v. 7, p. 2541.
MacLean, W.H., and Kranidiotis, P., 1987, Immobile elements as monitors
of mass transfer in hydrothermal alteration: Phelps Dodge massive sulfide
deposit, Matagami, Quebec: Economic Geology, v. 82, p. 951962.
Maske, S., and Skinner, B.J., 1971, Studies of the sulfosalts of copper: I.
Phases and phase relations in the system Cu-As-S: Economic Geology,
v. 66, p. 901918.
Mavrogenes, J., Henley, R.W., Reyes, A.G., and Berger, B., 2010, Sulfosalt
melts: Evidence of high-temperature vapor transport of metals in the for-
mation of high-sulfidation lode gold deposits: Economic Geology, v. 105,
p. 257262.
Meyer, C., and Hemley, J., 1967, Wall rock alteration, in Barnes, H.L., ed.,
Geochemistry of hydrothermal ore deposits: New York, Holt, Rinehart, and
Winston, p. 166234.
Michard, A., 1989, Rare earth element systematics in hydrothermal fluids:
Geochimica et Cosmochimica Acta, v. 53, p. 745750.
Migdisov, A.A., and Williams-Jones, A.E., 2013, A predictive model for trans-
port of silver chloride by aqueous vapor in ore-forming magmatic-hydrother-
mal systems: Geochimica et Cosmochimica Actamica, v. 104, p. 123135.
Migdisov, A.A., Williams-Jones, A.E., and Suleimenov, O.M., 1999, Solubility
of chlorargyrite (AgCl) in water vapor at elevated temperatures and pres-
sures: Geochimica et Cosmochimica Acta, v. 63, p. 38173827.
Migdisov, A.A., Williams-Jones, A.E., and Wagner, T., 2009, An experimental
study of the solubility and speciation of the rare earth elements (III) in fluo-
ride- and chloride-bearing aqueous solutions at temperatures up to 300C:
Geochimica et Cosmochimica Acta, v. 73, p. 70877109.
Miller, S.E., Heath, G.R., and Gonzalez, R.D., 1982, Effects of temperature
on the sorption of lanthanides by montmorillonite: Clays and Clay Minerals,
v. 30, p. 111122.
Moritz, R., Kouzmanov, K., and Petrunov, R., 2004, Late Cretaceous Cu-Au
epithermal deposits of the Panagyurishte district, Srednogorie zone, Bul-
garia: Swiss Bulletin of Mineralogy and Petrology, v. 84, p. 7999.
Morrice, M.G., and Gill, J.B., 1986, Spatial patterns in the mineraloogy of
island arc magma series: Sangihe Arc, Indonesia: Journal of Volcanology
and Geothermal Research, v. 29, p. 311353.
Morrice, M.G., Jezek, P.A., Gill, J.B., Whitford, D.J., and Monoarfa, M.,
1983, An introduction to the Sangihe Arc: Volcanism accompanying arc-arc
collision in the Molucca Sea, Indonesia: Journal of Volcanology and Geo-
thermal Research, v. 19, p. 135165.
Mountain, B.W., and Seward, T.M., 2003, Hydrosulfide/sulfide complexes of
copper (I): Experimental confirmation of the stoichiometry and stability of
Cu (HS)2 to elevated temperatures: Geochimica et Cosmochimica Acta, v.
67, p. 30053014.
Muntean, J.L., Kesler, S.E., and Russell, N., 1990, Evolution of the Monte
Negro acid sulfate Au-Ag deposit, Pueblo Viejo, Dominican Republic:
Important factors in grade development: Economic Geology, v. 85, p.
17381758.
Oberthr, T., Weiser, T., Amanor, J., and Chryssoulis, S., 1997, Mineralogi-
cal setting and distribution of gold in quartz veins and sulfide ores of the
Ashanti mine and other deposits in the Ashanti belt of Ghana: Genetic
implications: Mineralium Deposita, v. 32, p. 215.
Pacevski, A., Libowitzky, E., Zivkovic, P., Dimitrijevic, R., and Cvetkovic, L.,
2008, Copper-bearing pyrite from the Coka Marin polymetallic deposit,
Serbia: Mineral inclusions or true solid-solution?: The Canadian Mineralo-
gist, v. 46, p. 249261.
Pals, D.W., Spry, P.G., and Chryssoulis, S.L., 2003, Invisible gold and tel-
lurium in arsenic-rich pyrite from the Emperor gold deposit, Fiji: Implica-
tions for gold distribution and deposition: Economic Geology, v. 98, p.
479493.
 HIGH-SULFIDATION EPITHERMAL PYRITE-HOSTED Au ORE FORMATION, SANGIHE ISLAND, INDONESIA 1733
Posfai, M., and Buseck, P.R., 1998, Relationships between microstructure
and composition in enargite and luzonite: The American Mineralogist, v.
83, p. 373382.
Radcliff, D., and McSween, H.T., 1970, Copper zoning in pyrite from Cerro
de Pasco, Peru: Reply: The American Mineralogist, v. 55, p. 527528.
Reich, M.R., Kesler, S.E., Utsunomiya, S., Palenik, C.S., Chryssoulis, S.L.,
and Ewing, R., 2005, Solubility of gold in arsenian pyrite: Geochimica et
Cosmochimica Acta, v. 69, p. 27812796.
Robie, R.A., and Hemingway, B.S., 1995, Thermodynamic properties of min-
erals and related substances at 298.15 K and 1 Bar (105 Pascals) pressure
and at higher temperatures: U.S. Geological Survey Bulletin 2131, p. 461.
Rye, R.O., 1993, The evolution of magmatic fluids in the epithermal envi-
ronment: The stable isotope perspective: Economic Geology, v. 88, p.
733752.
2005, A review of the stable-isotope geochemistry of sulfate minerals in
selected igneous environments and related hydrothermal systems: Chemi-
cal Geology, v. 215, p. 536.
Rye, R.O., Bethke, P.M., and Wasserman, M.D., 1992, The stable isotope
geochemistry of acid sulfate alteration: Economic Geology, v. 87, p.
225262.
Salan, A., Villemant, B., Grard, M., Komorowski, J.-C., and Michel, A.,
2011, Hydrothermal alteration in andesitic volcanoes: Trace element
redistribution in active and ancient hydrothermal systems of Guadeloupe
(Lesser Antilles): Journal of Geochemical Exploration, v. 111, p. 5983.
Salvi, S., Pokrovski, G.S., and Schott, J., 1998, Experimental investigation of
aluminum-silica aqueous complexing at 300C: Chemical Geology, v. 151,
p. 5167.
Scher, S., Williams-Jones, A.E., and Williams-Jones, G., 2013, Fumarolic
activity, acid-sulfate alteration, and high sulfidation epithermal precious
metal mineralization in the crater of Kawah Ijen volcano, Java, Indonesia:
Economic Geology, v. 108, p. 10991118.
Seo, J.H., Guillong, M., and Heinrich, C.A., 2009, The role of sulfur in the
formation of magmatichydrothermal coppergold deposits: Earth and
Planetary Science Letters, v. 282, p. 323328.
Sharman, E.R., 2011, Application of multiple sulfur isotope analysis to
Archean ore-forming processes: Unpublished Ph.D. Thesis, Montral,
Canada, McGill Univesity, 196 p.
Simon, G., Huang, H., Penner-Hahn, J.E., Kesler, S.E., and Kao, L.-S., 1999,
Oxidation state of gold and arsenic in gold-bearing arsenian pyrite: Ameri-
can Mineralogist, v. 84, p. 10711079.
Springer, G., 1969, Compositional variations in enargite and luzonite: Min-
eralium Deposita, v. 4, p. 7274.
Stefnsson, A., and Seward, T.M., 2004, Gold(I) complexing in aqueous sul-
phide solutions to 500C at 500 bar: Geochimica et Cosmochimica Acta, v.
68, p. 41214143.
Stoffregen, R.E., 1987, Genesis of acid-sulfate alteration and Au-Cu-Ag
mineralization at Summitville, Colorado: Economic Geology, v. 82, p.
15751591.
Stoffregen, R.E., and Alpers, C.N., 1987, Woodhouseite and svanbergite in
hydrothermal ore deposits: Products of apatite destruction during advanced
argillic alteration: The Canadian Mineralogist, v. 25, p. 201211.
Stoffregen, R.E., and Cygan, G.L., 1990, An experimental study of Na-K
exchange between alunite and aqueous sulfate solutions: The American
Mineralogist, v. 75, p. 209220.
Stoffregen, R.E., Alpers, C.N., and Jambor, J.L., 2000, Alunite-jarosite crys-
tallography, thermodynamics, and geochronology: Reviews in Mineralogy
and Geochemistry, v. 40, p. 453479.
Stone, M., 2010, Sangihe property, Sangihe Island, North Sulawesi, Indone-
sia: Independent technical report, 130 p.
Taylor, B.E., 2007, Epithermal gold deposits, in Goodfellow, W.D. ed., Min-
eral deposits of Canada: A synthesis of major deposit-types, district metal-
logeny, the evolution of geological provinces, and exploration methods:
Geological Association of Canada, Mineral Deposits Division, St. Johns,
p. 113139.
Thiersch, P.C., Williams-Jones, A.E., and Clark, J.R., 1997, Epithermal min-
eralization and ore controls of the Shasta Au-Ag deposit, Toodoggone Dis-
trict, British Columbia, Canada: Mineralium Deposita, v. 32, p. 4457.
Thode, H.G., Monster, J., and Dunford, H.B., 1961, Sulphur isotope geo-
chemistry: Geochimica et Cosmochimica Acta, v. 25, p. 150174.
Thode, H.G., and Rees, C., 1979, Sulphur isotopes in lunar and meteorite
samples: Proceedings of Lunar Planetary Science Conference 10th, Hous-
ton, Texas, March 1317, 1979, p. 16291636.
Ulrich, T., Gunther, D., and Heinrich, C.A., 1999, Gold concentrations
of magmatic brines and the metal budget of porphyry copper deposits:
Nature, v. 399, p. 676679.
van Hinsberg, V.J., and Schumacher, J.C., 2007, Intersector element parti-
tioning in tourmaline: A potentially powerful single crystal thermometer:
Contributions to Mineralogy and Petrology, v. 153, p. 289301.
Voudouris, P.C., 2010, Conditions of formation of the Mavrokoryfi high-
sulfidation epithermal Cu-Ag-Au-Te mineralization (Petrota Graben, NE
Greece): Mineralogy and Petrology, v. 101, p. 97113.
Warren, I., Simmons, S.F., and Mauk, J.L., 2007, Whole-rock geochemical
techniques for evaluating hydrothermal alteration, mass changes, and com-
positional gradients associated with epithermal Au-Ag mineralization: Eco-
nomic Geology, v. 102, p. 923948.
White, D., 1957, Thermal waters of volcanic origin: Geological Society of
America Bulletin, v. 68, p. 16371657.
White, N.C., and Hedenquist, J.W., 1990, Epithermal environments and
styles of mineralization: Variations and their causes, and guidelines for
exploration: Journal of Geochemical Exploration, v. 36, p. 445474.
Wilder, A.M., and Seward, T.M., 2002, The adsorption of gold(I) hydrosulfide
complexes by iron sufide surfaces: Geochimica et Cosmochimica Acta, v.
66, p. 383402.
Williams-Jones, A.E., and Heinrich, C.A., 2005, Vapor transport of metals
and the formation of magmatic-hydrothermal ore deposits: Economic
Geology, v. 100, p. 12871312.
Williams-Jones, A.E., Migdisov, A.A., Archibald, S.A., and Xiao, Z., 2002,
Vapor-transport of ore metals, in Hellmann, R., and Wood, S.A. eds., Water-
rock interactions, ore deposits, and environmental geochemistry: A tribute
to David A. Crerar: The Geochemical Society, p. 279306.
Williams-Jones, A.E., Bowell, R.J., and Migdisov, A.A., 2009, Gold in solu-
tion: Elements, v. 5, p. 281287.
Zezin, D., Migdisov, A.A., and Williams-Jones, A.E., 2011a, PVTx proper-
ties of H2O-H2S fluid mixtures at elevated temperature and pressure based
on new experimental data: Geochimica et Cosmochimica Acta, v. 75, p.
54835495.
2011b, The solubility of gold in H2O-H2S vapour at elevated tempera-
ture and pressure: Geochimica et Cosmochimica Acta, v. 75, p. 51405153.
Zhang, X., 1993, SULF34: A QUICKBASIC program for calculating sulfur
isotope distribution in log fO2-pH diagram: Computers & Geosciences, v.
19, p. 817823.
Zotov, N., Petrov, K., and Dimitrova-Pankova, M., 1990, Infrared spectra of
Cu (II)-Co (II) mixed hydroxide nitrates: Journal of Physics and Chemistry
of Solids, v. 51, p. 11991205.