The Bingham Canyon Porphyry Cu-Mo-Au Deposit. III. Zoned Copper-Gold Ore Deposition by Magmatic Vapor Expansion °

2010 Society of Economic Geologists, Inc.
Economic Geology, v. 105, pp. 91118
The Bingham Canyon Porphyry Cu-Mo-Au Deposit.

III. Zoned Copper-Gold Ore Deposition by Magmatic Vapor Expansion
MARIANNE R. LANDTWING,1,* CAROLINE FURRER,1,** PATRICK B. REDMOND,2, THOMAS PETTKE,1,
MARCEL GUILLONG,1, AND CHRISTOPH A. HEINRICH3,
1ETH Zurich, Department of Earth Sciences, Clausiusstrasse 25, 8092 Zurich, Switzerland
2Department of Geological and Environmental Sciences, Stanford University, Stanford, California 94305
3ETH Zurich, Department of Earth Sciences, Clausiusstrasse 25, 8092 Zurich, Switzerland, and
Faculty of Mathematics and Natural Sciences, University of Zurich
Abstract
Fluid inclusion microthermometry and laser-ablation ICPMS microanalysis are combined with geological
and textural observations to reconstruct the spatial and temporal evolution of magmatic fluids that formed the
subvolcanic porphyry Cu-Au(-Mo) ore deposit at Bingham Canyon, Utah. The Bingham Canyon orebody is
exposed over ~1.6 km vertically and has the shape of an inverted cup with distinct metal zoning.
Fluid inclusions in the barren but highly veined and potassically altered deep center of the system have
intermediate density (~0.6 g cm3) and a salinity of ~7 wt percent NaCl equiv. They have subequal concentra-
tions of Na, K, Fe, and Cu and contain minor CO2. The intermediate-density fluids were trapped as a single
phase, mostly at >500C and >800 bars. The Au-Curich center near the top of the orebody contains low-
density vapor inclusions (~0.2 g cm3) coexisting with brine inclusions containing ~45 wt percent NaCl equiv.
The vertical transition of different inclusion types indicates phase separation of the single-phase input fluid
upon volume expansion associated with a pressure drop to 200 100 bars. Mass-balance calculation based on
all analyzed inclusion components indicates that the mass of the vapor phase exceeded that of the brine by
~9/1. The vapor contained Cu as its dominant cation (~1.5 wt %) and contributed about 95 percent of the total
amount of copper transported to the base of the orebody. Bornite, chalcopyrite, and native gold were precipi-
tated in a narrow temperature interval from 430 to 350C, into secondary pore space created by local redis-
solution of vein quartz as a result of retrograde quartz solubility in the vapor-dominated fluid system.
Intermediate-density fluid inclusions in the deepest parts of the peripheral copper ore zone have identical
density and composition, including similar gold contents, as those in the deep center. Microthermometry and
statistical estimation of phase proportions in the inclusions show that the vapor in the peripheral Cu-rich but
Au-poor ore zone remained denser, and the separating brine was less saline (~36 wt % NaCl equiv), compared
to vapor and brine in the central Au-Cu ore zone. This indicates that the peripheral fluids experienced a lower
degree of phase separation, due to slightly higher fluid pressure at equivalent temperature, compared to more
strongly expanding fluids in the center of the system.
The systematic zoning of Au/Cu within the ore shell, despite compositionally similar input fluids, is inter-
preted to have resulted from slightly different pressure-temperature-density evolution paths of magmatic flu-
ids. Copper was selectively precipitated in the peripheral ore zone, in contrast to complete coprecipitation of
Au and Cu in the central upflow zone of the vapor plume. The formation of particularly rich Cu-Au ore in the
center of the upward-expanding fluid plume is consistent with published experimental data, showing that the
solubility of metals in hydrous vapor decreases sharply with falling pressure, due to destabilization of the
hydration shell around metal complexes in expanding vapor. This interpretation supports the classic vapor
plume model for porphyry copper ore formation but additionally emphasizes the role of sulfur-bearing com-
plexes as a key chemical control on magmatic-hydrothermal metal transport and the deposition of Cu and Au
in porphyry ores. Our interpretation of selective Cu Au precipitation as a function of vapor density can
explain the more general observation that most gold-rich porphyry copper deposits are formed in shallow sub-
volcanic environments, whereas deeper seated porphyry Cu-(Mo) deposits are generally gold poor.
Introduction quantitative, integrating the role of magmatic fluids exsolved

INTERPRETATIONS of porphyry-style ore formation over the from plutons underlying the mineralized stocks (e.g., Henley
last decades have become increasingly differentiated and and McNabb, 1978) with the convection of external fluids dri-
ven by the thermal energy of magmatic intrusions (Taylor,
*Present address: University of Teacher Education Central Switzerland 1974; Cathles, 1977), as reviewed by Hedenquist and Lowen-
(PHZ Luzern), Museggstrasse, Lucerne, Switzerland. stern (1994), Seedorff et al. (2005), and Sillitoe (2010). Early
**Present address: Burstwiesenstrasse 53, 8606 Greifensee, Switzerland.
Teck Resources Limited, 550 Burrard Street, Vancouver, BC Canada V6C studies of fluid inclu-
sions in porphyry cop- Acronyms Used in this Paper
0B3.
Present address: University of Bern, Baltzerstrasse, Bern, Switzerland. per deposits recognized BP = biotite prophyry
Present address: University of Tasmania, CODES, Private Bag 79, Ho-
two locally coexisting EDM = early dark micaceous
bart, Australia 7001. fluid phases, a high- LP = latite porphyry
Corresponding author: e-mail, heinrich@erdw.ethz.ch
salinity brine and a low- QLP = quartz latite porphyry
A digital supplement to this paper is available at <http://www.geoscience QMP = quartz monzonite porphyry
world.org/> or, for subscribers, on the SEG website. salinity vapor (Roedder,
Submitted: May 13, 2009
0361-0128/10/3866/91-28 91 Accepted: December 29, 2009
92 LANDTWING ET AL.
1971, 1992; Denis et al., 1980). Magmatic brine inclusions documented in five research theses at Stanford University
contain high ore metal concentrations, based on observations and ETH Zurich by Inan (2002), Redmond (2002),
of chalcopyrite daughter crystals (e.g., Roedder, 1971; Nash, Landtwing (2004), Furrer (2006), and Gruen (2007).
1976; Denis et al., 1980), and experimental data indicated
that chloride complexing allows transfer of copper from hy- Mine-Scale Geology
drous magmas to saline magmatic fluids (Candela and Hol- The Bingham Canyon deposit is hosted by the Bingham
land, 1986; Candela and Piccoli, 1995). These observations stock, a polyphase subvolcanic intrusion of broadly monzonitic
led to the interpretation that chloride in saline magmatic flu- composition emplaced into Paleozoic quartzites, semipelitic
ids was the key ingredient in the formation of porphyry cop- siltstones, and impure carbonate rocks (Fig. 1). The sedimen-
per deposits (Lindgren, 1905; Dilles, 1987; Cline and Bodnar, tary rocks were folded and thrusted during several phases of
1991). Other studies suggested that the low-salinity magmatic Mesozoic orogeny, prior to porphyry intrusion and hydrother-
vapor phase is the main fluid driving porphyry-style ore for- mal mineralization in the late Eocene. Later block faulting by
mation (Henley and McNabb, 1978; Giggenbach, 1992), con- Basin-and-Range extension during the Oligocene to Pliocene
sistent with early evidence indicating high Cu contents in led to 10 to 30 tilting and exposure of the deposit in the
low-salinity vapor inclusions (Etminan, 1977; Eastoe, 1978; Oquirrh Mountains, west of Salt Lake City. This tilting pre-
Sawkins and Scherkenbach, 1981; Lowenstern et al., 1991). served the southeastern flank of a former stratovolcano of co-
Analytical studies of fluid inclusions from porphyry-style magmatic (broadly latitic) lavas and volcaniclastic rocks (Deino
ore deposits and miarolitic cavities in barren and mineralized and Keith, 1997; Waite et al., 1997). Geologic relationships al-
intrusions have quantified ore metal concentrations in a wide lowed an approximate reconstruction of the paleotopography
range of magmatic-hydrothermal fluids, showing that both at the time of volcanism and magmatic-hydrothermal mineral-
phases can carry high concentrations of ore metals and sulfur ization, indicating ~2 km of rock overburden above the pre-
(Audtat et al., 2008). Some of the highest copper and gold mine top of the deposit (Deino and Keith, 1997).
concentrations are recorded by low-salinity fluids of vaporlike The Bingham stock (Fig. 1) has a surface extent of ~3 3
density (Heinrich et al., 1992, 1999; Damman et al., 1996; km and comprises at least five intrusive phases (Redmond and
Audtat et al., 1998; Ulrich et al., 1999; Baker et al., 2004; Einaudi, 2010), starting with the premineralization equigranu-
Klemm et al., 2007). These observations have spurred a new lar monzonite and several nonmineralizing dikes that intruded
wave of experimental and conceptual studies, which demon- it. The quartz monzonite porphyry (QMP) is a medium- to
strate that hydrous magmatic vapor can transport high con- coarse-grained elongate stock, measuring about 1.5 0.4 km
centrations of ore metals (e.g., Williams et al., 1995; in plan, which intruded the more mafic equigranular mon-
Archibald et al., 2002; Williams-Jones et al., 2002; Williams- zonite along its northwest contact. The QMP was the focus of
Jones and Heinrich, 2005; Pokrovski et al., 2005, 2008; Simon greatest intensity of veining, pervasive potassic alteration, and
et al., 2007; Nagaseki and Hayashi, 2008; Zezin et al., 2008). high-grade Cu-Au mineralization, which extended into the ad-
This paper investigates the evolution of hydrothermal fluids jacent equigranular monzonite and (to a lesser extent) into the
of variable density and salinity in the Bingham Canyon Cu- sedimentary wall rocks (Fig. 2). Thin (generally <50 m; Fig. 1)
Mo-Au deposit (Fig. 1), based on detailed microanalytical but more laterally extensive dikes of latite porphyry (LP) in-
study of fluid inclusions. Bingham Canyon is an exceptionally truded later along the same axis as the QMP (Fig. 1). Although
large and rich deposit but exhibits geologic features that are the LP dikes locally have a lower vein density and lower grade
typical for many other porphyry-style ore deposits. This in- mineralization (Redmond and Einaudi, 2010), they have un-
cludes a prominent metal zonation centered on an intrusive dergone pervasive biotite alteration of the groundmass and
stock (see Gruen et al., 2010; Redmond and Einaudi, 2010). mafic phenocrysts. At the mine scale, Cu-Au distribution in the
The ore shell has the shape of an inverted cup and is well ex- LP follows the same concentric pattern as that in the QMP, and
posed in the large open-pit mine and by deep diamond drill the mineralized veinlets are indistinguishable except by trun-
holes that have reached the base of the zoned orebody and its cations seen at some intrusive contacts. A fourth group of thin
barren core, more than a kilometer below the center of the dikes include dark brown biotite porphryry (BP) of lampro-
deposit (Fig. 2). This vertical exposure provides an opportu- phyric texture and composition and quartz latite porphyry
nity to reconstruct the fluid evolution, both temporally in the (QLP), which intruded all earlier porphyries (mostly <20 m
evolving magmatic-hydrothermal system, as well as spatially thickness). Both of the dike types were cut by minor stockwork
from the deep region of fluid ascent to the region of zoned veinlets with restricted biotite alteration.
ore mineral precipitation. Fluid inclusions at Bingham The sequence of vein formation was independently recorded
Canyon are large and well preserved, suitable for pioneering in numerous observation points throughout the deposit, based
technical and analytical studies (Roedder, 1971; Anderson et on intersection and overprinting relationships between veins,
al., 1989; Irwin and Roedder, 1995; Bodnar, 1995; Kendrick intrusive contacts, and different alteration types (Redmond et
et al., 2001a, b; Vanko et al., 2001; Redmond et al., 2004, al., 2001; Redmond, 2002; Landtwing, 2004; Redmond and
Landtwing et al., 2005; Guillong et al., 2008). This fluid in- Einaudi, 2010; Gruen et al., 2010; Fig. 3). Wispy biotite vein-
clusion study builds on two companion papers documenting lets and early dark micaceous (EDM) veins were the first, vol-
the intrusive history and paragenesis (Redmond and Einaudi, umetrically minor, veins that formed after each of the miner-
2010) and the significance of intrusion and vein geometry, in- alized intrusions. The EDM veins are marked by 1- to
dicating that the structure of the ore shell resulted from radial 2-cm-wide alteration halos, which locally have a thin (<3 mm)
extension above a broad high-pressure fluid source (Gruen et centerline of quartz. This quartz may have been deposited
al., 2010). Detailed observations for this series of papers are after reopening during the subsequent stockwork vein stage,
0361-0128/98/000/000-00 $6.00 92
BINGHAM CANYON: ZONED Cu-Au ORE DEPOSITION BY MAGMATIC VAPOR EXPANSION 93
3
9
1
5
IGNEOUS ROCK TYPES (Eocene)
14
31
0 1000 2000 feet
tic N
True N
Quartz latite porphyry
0 300 600 meters
ne
Mi
Magnetic
ne
Mag
Latite porphyry
N
declination:
Quartz monzonite porphyry Cross section at 1400W
Hybrid quartz monzonite porphyry 0.30 ppm Au grade
Porphyritic quartz monzonite 0.35 wt.-% Cu grade
Equigranular monzonite
SEDIMENTARY ROCK TYPES

(Pennsylvanian) OQUIRRH GROUP
Calcareous sandstone
Quartzite and
calcareous quartzite A
Limestone and skarn
Endoskarn
KE
Igneous breccia
DI
E
Y DIK
SS
AR
LE AND
FOR
ST
TUN
AS
ILL
PAR
N ELL
B EDS
DS
BE
E LL BINGHAM
RN
PA STOCK
E
IK
D
LL
HI
N
AI
M CO
MM
ER
CI
NE
AL
LIM OHIO
O E
ES
T
IK ES
TO COPPER
LI
M
D YD NE
DIKE
AL AN
CI
E
ER
IK
M
ILL
D
M
CO IN H
MA
NE
TO
ES
M
LI
AN
RD
JO
JOR
DAN
LIM
KE
PHOENIX DI
E STO
NE
B
LAST CHANCE
STOCK
JORDAN
Bingham UTAH
LIMESTONE Canyon
FIG. 1. Mine-scale map based on Kennecott Utah Copper, showing major lithologic units and ore-grade contours of >0.35
percent Cu and >0.3 percent Au before the pit surface intersected the barren core (mapping stage of 1998; compare with
Gruen et al., 2010, fig. 2). Line A-B shows trace of cross section of Figures 3 and 7.
because the EDM veins formed at quartz-undersaturated con- alteration. Truncations by intrusive contacts demonstrate that
ditions, as indicated by the typical alteration assemblage of quartz stockwork veins formed after solidification of each of
sericite + biotite + K-feldspar andalusite corundum Cu- the three mineralized intrusions (QMP, LP, and QLP) but
Fe sulfides. Quartz stockwork veins comprise a complex set with decreasing abundance and intensity of associated potas-
of crosscutting veinlets. They are dominated by quartz infill sic alteration halos. The quartz stockwork veins (Fig. 3A, B)
and are surrounded by variably pervasive biotite-K-feldspar are typically 0.5 to 2 cm thick and have variable appearances,
0361-0128/98/000/000-00 $6.00 93
94 LANDTWING ET AL.
D195
D404
D211
D123
D109
D420
D430
D188
D169
D74
A57
A51
NW SE
2400 m
4000 N
2000 N
0000 N
4000 S
6000 S
2000 S
asl.
B
A57
A ph y
g ra
A51
p o
g to
D74
Pi asl.
to n in
D169
2000 m
D123
ut l
mi
in e-
D188
20 Pr
e
6000 ft 06 224
D195
Au-Cu center
D109
D211
D420
582
D430
19
D404
176 Cu-rich periphery 1600 m
Sedimentary 221
host rock 29 1587
1719
888 1277
1032
1039
1386
4000 ft 1200 m
Below orebody
1632
Sedimentary
host rock
800 m
100 m
2000 ft 4788
Ore grades
3657
Deep center 3749 4491 ~0.15 ppm Au
5889.5 400 m
5910
Deep periphery ~0.30 ppm Au
6063 Quartz ~1.00 ppm Au
Monzonite
Equigranular >0.15 wt.-% Cu
Porphyry
Monzonite
>0.35 wt.-% Cu
5192 >0.70 wt.-% Cu
0000 ft 000 m
FIG. 2. Northwest-southeast cross section 1400E of the mine grid, perpendicular to the elongate QMP intrusion on
which the zoned Cu-Au mineralization is centered. Note highest Au-Cu grade in the top of inverted cup-shaped orebody,
and the Cu-rich but Au-poor limbs extending downward at the periphery of the system. Mo contours (not shown) largely fol-
low Cu contours but tend to be slightly displaced downward toward the barren center below the Au-Curich orebody (Gruen
et al., 2010). Grid in mine coordinates (ft). Sampling regions and sample points in drill holes and surface projection (Table
1) are indicated, black dots denoting quartz stockwork veins analyzed by LA-ICPMS, black triangles are quartz-molybden-
ite veins analyzed by LA-ICPMS, and white dots are quartz stockwork veins studied by fluid inclusion statistics. Two addi-
tional stockwork vein samples measured by microthermometry by Redmond (2002) are marked with black stars. Sample
D411-2646.5 (off section) and surface samples 4990-2050, 5090-1510, 5090-1525, and 5190-1600 (collected near the collar
of drill hole 211) are not included in this Figure.
from early irregular-walled glassy or granular quartz veins to mapping and core logging to document the time relationships
more straight-walled veins with symmetric columnar quartz of major vein types and the contained minerals. Figure 2 and
infill. They generally have no open spaces, and only some Table 1 define five sampling regions with regard to simplified
show a well-defined centerline. In contrast to the multiple geology (Redmond and Einaudi, 2010) and large-scale Cu
phases of stockwork veins, quartz-molybdenite veins postdate and Au grade distribution (Gruen et al., 2010). More than half
all intrusions including QLP (Fig. 3B). The veins are 1 to 5 of the samples from the open-pit surface and from drill cores
cm wide, typically have at least one meter of lateral extent, were taken in the vicinity of section 1400E, i.e., the main
and consist of blocky to euhedral quartz and coarse flakes of cross section discussed in the two preceding papers. The
molybdenite that grew from straight walls into still partially focus of this paper is on the quartz stockwork veins, which are
open space. Quartz-pyrite veins with sericitic alteration are intimately linked to the zoned Cu and Au mineralization, and
straight-sided to anastomosing and occur as vertically exten- with potassic alteration extending into the deep barren center
sive swarms of thin (< 1 cm) fractures filled by pyrite, minor of the ore system. Most samples are hosted by QMP and EM
quartz calcite chalcopyrite bornite. The quartz-pyrite away from contacts with later intrusions. This sampling strat-
veins with sericitic alteration are surrounded by halos of a few egy avoids postentrapment modification of inclusions by re-
centimeters to several meters extent, in which all feldspars heating due to local intrusions but introduces the uncertainty
and most mafic minerals have been altered to sericite and/or that the fluids cannot be uniquely associated with the main
illite, pyrite, and in some cases chlorite (Fig. 3C). mineralizing QMP phase, although this is statistically most
likely in light of the decreasing vein density associated with
Methods later and volumetrically minor dikes. We have not selectively
The fluid inclusion study reported here is based on approx- sampled the volumetrically minor quartz stockwork veins cut-
imately 500 samples, which were collected during systematic ting the later porphyry dikes (Redmond and Einaudi, 2010),
0361-0128/98/000/000-00 $6.00 94
which is a limitation of the present study. However, a few

A quartz-molybdenite veins were sampled preferably where
there was direct evidence that the vein has cut by a late BP or
QLP dike.
Several hundred vein samples were prepared as quickplates
for inspection of inclusion type and quality, i.e., showing the
largest and best preserved inclusions in clearly identifiable as-
semblages (spatially associated groups of coeval inclusions
with similar petrographic appearance and constant phase pro-
portions at room temperature; Goldstein and Reynolds,
1994). A subset of 80 sections was polished and studied by
conventional reflected and transmitted light microscopy.
Cathodoluminescence imaging on a scanning electron micro-
1 cm scope (SEM-CL) of the majority of these samples revealed at
least two quartz generations in each vein, which could be cor-
related with the history of fluid inclusion entrapment in se-
B lected quartz-Cu-Fe sulfide veins (Redmond et al., 2004;
Landwing et al., 2005). As detailed in Table 1, a selection of
27 doubly polished sections were studied by a combination of
1 cm cathodoluminescence petrography, microthermometry, laser
ablation inductively coupled plasma mass spectrometry (LA-
ICPMS) and/or statistical evaluation of phase proportions at
room temperature. All microthermometric and LA-ICPMS
data were collected from petrographically defined inclusion
assemblages rather than individual inclusions to allow calcu-
lation of averages and standard deviations (1) whenever sev-
eral measurements of a particular property or analytical con-
centration ratio was obtained from individual inclusions in the
assemblage. All analytical data points in this paper refer to
such assemblages, as tabulated in the digital supplement. This
approach ensures that analytical errors or uncertainties due to
imperfect trapping or inclusion preservation (variability
within one assemblage) can be distinguished from true geo-
logic variation in paleofluid properties (differences between
assemblages; Goldstein and Reynolds, 1994; Heinrich et al.,
2003). Relating such variations to geologic time, however, is
generally not possible because the majority of identifiable and
measureable fluid inclusion assemblages are secondary or
pseudosecondary in origin. Unambiguously primary inclu-
sions do occur as faint brownish-colored crystal growth zones
C in many veins, but they are invariably too small for quantita-
2 cm tive analysis (<2 m) or even for phase type identification.
Mapping of fluid inclusion characteristics identified a sys-
tematic variation in predominant inclusion types with regard
to depth (Redmond et al., 2004) and also indicated that phase
proportions in the brine and vapor inclusions varied system-
atically from the Au-Curich center of the orebody to the Au-
poor peripheral Cu ore zone of the system (Fig. 2). Five of
the samples described in Table 1 were studied with a previ-
ously untested statistical approach; two from the Cu-rich pe-
riphery of the orebody and three samples from the center of
the system in and below the high-grade Au-Cu orebody. In
each petrographic section, two- and three-phase inclusions
were counted along random straight lines as indicated in Fig-
FIG. 3. Photographs of slabs of the three main vein types studied by fluid ure 4. Inclusions with a minimum dimension of 20 m and a
inclusion analysis. (A). Quartz stockwork veins in biotite-altered equigranular depth from the surface of 10 5 m (depth of focus) inter-
monzonite. (B). Quartz-molybdenite vein cutting biotite-altered LP, which sected by a reference line in the X-Y stage of the microscope
intruded QMP that had previously been Cu-Au mineralized by the main
stage of stockwork veining (lower part of photo). (C). Quartz-pyrite veins
were counted until exactly 60 two-phase inclusions (liquid +
with sericitic alteration cutting and overprinting earlier stockwork veins, as vapor small opaque crystal) or 60 three-phase inclusions
the last event of magmatic-hydrothermal mineralization. (liquid + vapor + halite other daughter crystals) were
0361-0128/98/000/000-00 $6.00 95
96 LANDTWING ET AL.
TABLE 1. Summary Description of Samples, Veins, and Fluid Inclusions
Sampling sites Samples: Vein, wall-rock, and alteration mineralogy; methods and data source
Main Vein minerals Vein

lithologic additional Vein width Wall
Location units Alteration Sample no. Vein type to qtz selvage (mm) rock
Deep QMP, Potassic A51-4788 Molybdenite veinlet reopening a Kspar 8 QMP

center (QLP, EM, Weak phyllic barren quartz stockwork vein
sedimen- overprint
tary rock) (argillization A57-5889.5 Quartz stockwork vein (white qtz) qtz, Kspar 9 QMP
of plag
phenocrysts)
A57-5910 Quartz molybdenite vein (fine- moly 8 QLP

grained, banded with undulating, (distinct
irregular borders, anastomosing) zones), cc,
A57-6063 Quartz stockwork vein (barren) zeo, Kspar 8 QMP
Center QMP, Potassic D109-1039 Quartz stockwork vein (barren, Kspar, cc White 6 QMP
below EM, Propyllitic parallel-walled, banded, blocky qtz) (in vugs), Kspar
orebody (LP, QLP) overprint minor cpy
of EM
Weak phyllic D195-1386(A) Quartz-molybdenite vein (gray, moly 10 LP
overprint clear qtz, 12 vol % idiomorphic (crystals up
(argillization moly flakes in distinct zones) to 5 mm)
of plag
phenocrysts) D404-1632 Two crosscutting quartz stockwork Kspar 4/2-7 QMP
veins (fine-grained, weakly zoned;
blocky, coarse-grained)
Au-Cu QMP, Potassic D195-888 Early dark micaceous vein (irregular, Kspar qtz, Kspar, 2-3 QMP
center, LP, QLP Weak phyllic anastomosing centerline, 15 mm mica, bn,
lower part overprint brown-green alteration selvage) cov, cpy
(argillization
of plag D195-1032 Quartz stockwork veins (two qtz Minor py 10 QMP
phenocysts) gen.: earlier milky phenocrysts,
reopening with clear, gray qtz)
D211-221 Quartz stockwork vein (vuggy, cpy, bn 8 QMP

clear qtz)
Au-Cu QMP, Potassic D123-224 Quartz stockwork vein cpy 10 QMP

center, LP, QLP Weak phyllic (clear to gray qtz)
upper part overprint
(argillization
of plag D211-19 Quartz stockwork vein (parallel- cpy, bn 12 QMP
phenocysts) walled, clear qtz)
D211-176 Quartz-molybdenite vein (sparry, moly 10 QMP

crosscutting gray quartz stockwork
vein with cpy, bn, Kspar)
D404-29 Quartz stockwork vein (clear qtz) bn, dig, cpy 9 QMP
4990-2050 Quartz-chalcopyrite-molybdenite moly 10 QMP

vein (sparry, gray, clear)
5090-1510 Quartz stockwork vein bn, dig, cpy 8 QMP

(clear to gray qtz)
0361-0128/98/000/000-00 $6.00 96
Observations, Ordered by Sampling Regions shown in Figure 3
Fluid inclusions: types, description and data extent
Sulfide Fluid # assem. # single FI # assem. # single FI

Alteration/matrix Measure- # CL with # inclusion Fluid inclusion description MT MT LA LA
mineralization ments Thesis gen. CL gen. types (dc = daughter crystal) ID/B/V/A ID/B/V/A ID/B/V/A ID/B/V/A
Potassic, weak SA/CL F/R ID,(B,V) ID: small inclusions,

phyllic op./barren irregularly to regularly shaped,
1 opaque dc (cpy)
Potassic, weak LA,MT,CL L 1 ID,(B,V) ID, V: up to 30 m, regular to ID18/ ID234/ ID17/ ID64/
phyllic op./barren negative crystal shaped/ID: B4/V4 B68/V19 B8/V6 B19/V15
two gen./B, V: postdating ID/
- B: regularly shaped, <15 m,
1 opaque dc
Potassic/barren LA,MT,CL L 1 ID,(B,V) ID, B, V: <15 m, irregular ID7/B2 ID59/B6 ID5/B3 ID18/B3
to regular shaped; B,
V: postdating ID; ID: 1
Potassic/barren MT,CL R ID opaque dc ID: containing CO2 ID1/B1 ID7/B1
Potassic, weak LA,MT,CL L 2 B,V,(A) B, V: regularly to negative B13 B307 B13/V4 B44/V4
phyllic op./barren crystal shaped, up to 30
m; B: 1 opaque dc, 1-2
transparent dc
Potassic/barren LA,MT,CL L 3 3 B,V,(A) B, V: irregularly to regularly B19 B366 B19/A1 B40/A1
shaped, up to 25 m; B: 2
opaque dc (cpy, moly),
2-3 transparent dc
Potassic/barren LA,MT/CL L/R 1/2 2 B,V,(A) B, V: 1 opaque daughter B10 B124 B6 B13
crystal, up to 30 m
Potassic/dissem- LA,MT,CL L 1 B,V,(A) B, V: regularly to negative B5 B59 B3 B8

inated cpy, bn crystal shaped, <15 m
Potassic, weak LA,MT/CL F/L 2 (2) B,V,(A) B: up to 40 m, 1 opaque dc B3/V3 B21/V20 B4/V4 B20/V25
phyllic op./minor (cpy), 2-3 transparent dc; V: up
disseminated cpy, bn to 40 m, 1 opaque dc (cpy)
Potassic, weak SA F B,V,(A) B, V: up to 45 m, irregularly

phyllic op./dissem- to regularly shaped; B: 1-2
inated cpy, bn opaque dc (cpy, hem), 2-4
transparent dc
Potassic/minor SA/CL F/L 2 2 B,V,(A) B, V: up to 50 m; B: 1-2

disseminated cpy opaque dc (cpy, hem),
2-4 transparent dc
Potassic/dissem- LA,MT,CL L 2 2 B,V,(A) B, V: two gen., second gen. B96/ B926/ B56/ B156/
inated cpy, bn associated with second gen. V16/A4 V97/A23 V32/A1 V79/A3
of qtz; B: 1-2 opaque dc,
1-3 transparent dc
Potassic/minor LA,MT,CL L 1 1 B,V,(A) B, V: multiple gen., B2/V5/A2 B26/ B3/V1 B5/V4

disseminated cpy, bn irregularly to regularly V37/A8
shaped, up to 15 m
Potassic/dissem- MT,CL R 2 2 B,V,(A) B, V: up to 50 m, two gen., B1 B9

inated cpy, bn second gen. associated with
copper deposition and
second gen. of qtz
Potassic/minor LA,MT L B,V,(A) B, V: multiple gen., B3/ B44/ B2/V1 B9/V5

disseminated cpy, bn irregularly to regularly V2/A2 V12/A8
shaped, up to 15 m; B: 1-2
opaque dc (cpy, hem),
2-3 transparent dc
Potassic/dissem- MT,CL R B,V,(A) B, V: irregularly to B1 B13

inated cpy, bn regularly shaped
0361-0128/98/000/000-00 $6.00 97
98 LANDTWING ET AL.
TABLE 1.
Sampling sites Samples: Vein, wall-rock, and alteration mineralogy; methods and data source
Main Vein minerals Vein

lithologic additional Vein width Wall
Location units Alteration Sample no. Vein type to qtz selvage (mm) rock
Au-Cu QMP, Potassic 5090-1525 Quartz stockwork vein cpy, bn, dig 9 QMP
center, LP, QLP Weak phyllic (two gen. of qtz)
upper part overprint
(argillization
of plag
phenocysts) 5190-1600 Quartz stockwork vein cpy, bn 10 LP
(gray, clear, sheeted)
Cu-rich EM Potassic D74-1719 Quartz stockwork vein (zoned, py, minor 15 EM

periphery Propyllitic reopening of clear qtz, cpy and cc
overprint abundant sulfides)
Phyllic
overprint D74-1587 Quartz stockwork vein (~15 vol % cpy as fractu- 9-11 EM
along cpy, three stages of qtz, gray (1st, re fillings and
cm-dm wide 70 vol %) and clear (2nd, 3rd) qtz) crystals, py
fractures
D169-582 Quartz stockwork vein cpy and cc 7 EM
D188-1277 Quartz stockwork vein (milky to py, cpy as 7 EM

clear qtz, sulfides in center of fracture
vein and as fracture fillings) fillings
D411-2646.5 Quartz-sericite-pyrite vein (py- py, cpy , bn, sericite 5 EM

qtz-bn-cpy, minor qtz, 8 mm to cov, enargite,
dm wide sericitic selvage) rutile
Deep EM Potassic D188-4491 Quartz stockwork vein (milky to py, minor cpy 15 EM
periphery Propyllitic clear qtz, with abundant sulfides)
overprint
D420-5192 Quartz stockwork vein (clear to py, minor cpy bio 5 EM

milky qtz with abundant sulfides)
D430-3657 Two crosscutting quartz stockwork py, cpy 3/4 EM

veins (clear, gray qtz) minor cc
D430-3749 Reopened quartz stockwork vein cpy, cc 5 EM

(blocky qtz)
Notes: Sample locations: the first digits of each drill core sample indicate the drill core number, whereas the second number gives the depths from the collar in
feet (e.g., sample D211-19: drill core D211 taken 19-ft depths below the collar); surface samples 4990-2050, 5090-1510, 5090-1525, and 5190-1600 are collected
near the collar of drill hole D211 (Fig. 3); sample D411-2646.5 is out of section originating from drill core collared at 325.6 E/2597.55 N at 5832.2-ft elevation
Abbreviations: General: assem. = assemblages; dc = daughter crystal; gen. = generation/generations; op. = overprint; FI = fluid inclusions
Measurements: SA = statistical analysis of distribution of inclusion types and phase proportions (cf. Fig. 4); LA = LA-ICPMS; MT = microthermometry;
CL = SEM-cathodoluminescence
reached. This always occurred first for the two-phase inclusions, of the three phases. These were plotted in frequency diagrams
providing a measure of the lower abundance of brine inclusions and converted to approximate volume proportions assuming a
seen in all samples. Each selected inclusion was photographed spherical vapor bubble and a halite cube inside a spherical in-
digitally through a 40 objective. For each inclusion, bubble, clusion. Even though imprecise for individual inclusions, the
liquid, and solids were hand-traced in Adobe Photoshop, and averaged area measurements yield a statistically unbiased and
areas were pixel-counted to give a measure of area proportions independent estimate of the average density of the two-phase
0361-0128/98/000/000-00 $6.00 98
(Cont.)
Fluid inclusions: types, description and data extent
Sulfide Fluid # assem. # single FI # assem. # single FI

Alteration/matrix Measure- # CL with # inclusion Fluid inclusion description MT MT LA LA
mineralization ments Thesis gen. CL gen. types (dc = daughter crystal) ID/B/V/A ID/B/V/A ID/B/V/A ID/B/V/A
Potassic/dissem- MT R B,V,(A) B, V: two gen., second gen. B1 B14

inated cpy, bn associated with second gen.
of qtz
Potassic/dissem- LA,MT L B,V,(A) B, V: up to 15 m, some B1/V2 B10/V13 B2/V1 B8/V3

inated cpy, bn inclusions trapped in apatite;
B: 1 opaque dc (cpy, hem),
1-2 transparent dc
Potassic, propylitic LA,MT F B,V,(A) B, V: up to 50 m; B: 1-2 B2/V8 B13/V36 B2/V4 B11/V21

op./disseminated transparent dc, 1 opaque dc
cpy, bn
Potassic/dissem- LA,MT,CL L 3 2 and 3 B,V,A B, V: irregularly to regularly B9/V3 B279/V11 B13 B37
inated cpy shaped, <20 m, clusters in
1st gen. of qtz, trails in 2nd
and 3rd but same chemical
composition; B: 1 opaque dc
Strong potassic/ LA,MT F ID,B,V ID: predominating, up to ID5 ID33 ID6 ID49
disseminated cpy 45 m; minor V, B, up to 45 m
Potassic/dissem- SA/CL F/L 2 2 B,V,ID B, V: predominating, up to

inated py, cpy 30 m; B: 1 opaque dc, 1-2
transparent dc; minor ID,
very small.
Potassic, phyllic LA,MT,CL L 3 1 py A A: CO2 bearing, <10 m, A19 A148 A5 A10

op./minor dissem- 2 cpy irregularly or regularly shaped,
inated cpy, bn some inclusions 1 opaque dc
Weak potassic, LA,MT F ID,(B,V) ID: predominating, up to ID4/B1 ID23/B4 ID4/B1 ID30/B5
weak propylitic op. 20 m; 1 opaque dc (hem)
disseminated py, cpy
Potassic, propylitic SA/CL F/L 1 1 ID,(B,V) ID: predominating, up to

op./abundant dissem- 30 m; minor V, B
inated py, cpy
Strong potassic/ LA,MT F ID,(V) ID: predominating, ID2 ID9 ID2 ID9
minor disseminated secondary, up to 25 m;
py, cpy minor V, 5-10 m
Potassic, weak LA,MT F ID,(B,V) ID: predominating, up to ID8/B1 ID37/B5 ID8 ID40
propylitic op./ 20 m, 1 opaque dc; minor
disseminated cpy V, B, small
Fluid inclusion types: ID = intermediate density; V = vapor; B = brine; A = aequeous liquid

Minerals: cc = calcite; Kspar = K-feldspar; plag = plagioclase; qtz = quartz; zeo = zeolite; bn = bornite; cov = covellite; cpy = chalcopyrite; dig = digen-
ite; hem = hematite; moly = molybdenite; py = pyrite
Thesis: F = Furrer; (2006); L = Landtwing (2004); R = Redmond (2002)
Full data for of 236 LA-ICPMS fluid inclusion assemblages (with 758 individual fluid inclusions measured) and for the distribution of fluid inclusion
types and phase proportions statistic samples are available as a digital supplement at <http://www.geoscienceworld.org/>
inclusions (ignoring CO2 and minor salt species) and of the and LA-ICPMS analysis. This corresponds to 758 individual
average total salinity of brines, assuming 26.5 percent NaCl in inclusions with 1 to 15 inclusions per assemblage; four addi-
the liquid phase (saturation concentration at 25C) with den- tional samples were studied by microthermometry only (see
sity 1.2 g cm3 and a halite density of 2.16 g cm3. right-hand columns in Table 1 for overview information and
In 18 samples, a total of 236 different inclusion assemblages digital supplement for all assemblage data). Low-temperature
were measured with a combination of microthermometry phase transitions and halite dissolution temperatures up to
0361-0128/98/000/000-00 $6.00 99
100 LANDTWING ET AL.
quartz vein tendency for overestimating the average concentration ratio

B1 in an assemblage due to the short signal duration (Pettke et
al., 2000). There is no possibility to quantify and correct this
potential error, but it does not affect the main conclusions
FIG. 4. Methodology for statistical from this study.
V2 analysis of phase proportions of vapor,
liquid, and salt crystals in fluid inclu-
sions of selected stockwork vein sam- Vein Petrography and Fluid Inclusion Results
V1 ples. For unbiased selection, every inclu-
sion cut by an arbitrary straight driving Quartz sulfide textures from SEM-CL
track (arrow) exceeding 20 m in size Quartz stockwork veins: Associated with Cu-Au mineraliza-
was counted for frequency estimation
of inclusion types and photographed tion these veins are generally massive and have a macroscop-
20m
for subsequent area measurement. ically glassy to granular appearance. Some contain a line of
small pink K-feldspar crystals growing on the vein wall or in-
side the vein. Others contain a calcite-filled centerline but no
~300C were usually measured first, whereas final homoge- macroscopic open spaces. In thin section and transmitted
nization experiments were done at the end on a few inclu- light, the quartz texture is mostly granular, but some brown
sions in each assemblage saved from destructive elemental growth bands consisting of submicroscopic fluid or melt in-
analysis. LA-ICPMS microanalysis on the prototype instru- clusions demonstrate precipitation into open space. Most
ment at ETH Zurich allows online petrographic control and quartz grains are anhedral and interlocked. Cathodolumines-
optimized quartz ablation (Gnther et al., 1997, 1998; Pettke cence (SEM-CL) imaging revealed a complex but consistent
et al., 2000). Note that LA-ICPMS measures ratios of ele- internal growth structure (Redmond et al., 2004; Landtwing
ment concentrations only. Determining absolute concentra- et al., 2005; Fig. 5). Similar to observations from other por-
tions requires a combination of this element ratio analysis phyry deposits (Rusk and Reed, 2002; Klemm et al., 2007,
with a concentration constraint from microthermometry, usu- 2008), the quartz stockwork veins at Bingham Canyon record
ally expressed as apparent salinity in wt percent NaCl equiv, two main stages of quartz crystallization (Fig. 5; Redmond et
and an assumption to estimate the true Na concentration by al., 2004). An early bright-luminescent quartz generation
combination with the LA-ICPMS analysis (see Heinrich et (Q1) constitutes the bulk of each quartz vein (>80 vol %). In
al., 2003, for calculation methods and discussion of uncer- deep barren stockwork veins, Q1 is homogeneously bright lu-
tainties). Abundant K, Fe, and Cu show that the Bingham minescent and fills most of the vein volume (Fig. 5A-C),
Canyon fluids strongly deviate from the binary NaCl-H2O whereas in the Cu Au mineralized ore zone it commonly
system. Moreover, apparent salinities of two-phase inclusions shows complex oscillatory zoning, from bright to gray CL in-
were difficult to quantify due to the presence of significant tensity (Fig. 5D-H). Intact Q1 grains were never observed to
CO2. Therefore, LA-ICPMS data plots and genetic interpre- contain inclusions of Cu-Fe sulfides or gold. A second, dull-
tations are primarily based on the directly measured element to nonluminescent quartz generation Q2 constitutes at most
ratios relative to Na (digital supplement). To obtain the rep- 20 vol percent of the veins. It occurs on cracks and grain
resentative fluid compositions shown in Tables 2 and 3, we boundaries that cut and overgrew Q1 in crystallographic con-
combined mass-balance constraints with the measurements tinuity and commonly truncated earlier growth bands in Q1
from microthermometry and LA-ICPMS analyses, as ex- (Fig 5D, G). Textures indicate that a process of microbreccia-
plained in the Discussion section below. Limits of detection tion and dissolution of Q1 occurred prior to recementation by
were calculated for each inclusion (Gnther et al., 1998; Q2. Chalcopyrite and composite bornite-digenite grains in-
Heinrich et al., 2003), which typically results in some values variably are in contact with and commonly surrounded by this
for trace components (notably Au) that are below the average dull-luminescent quartz Q2, which also extends along micro-
of detected values in the inclusion assemblage. This implies a cracks and grain boundaries into the potassically altered wall
FIG. 5. Characteristic textures in three vein types, as shown by SEM-CL (left), BSE images (center), and a small trans-
mitted light micrograph for overview (right). Note that CL brightness is adjusted for optimal contrast in each composite map
but not equivalent between the three samples shown. (A-C, sample A57-5889.5): Detail from three mutually intersecting
parallel-walled quartz stockwork veinlets from the deep center in barren but pervasively biotite-altered QMP, with granular
gray to milky macroscopic appearance; ~5 percent open space appears white in the BSE image (B) due to Ce-rich polishing
powder but may have originally been filled with calcite; CL image (A) shows poorly zoned anhedral quartz grains with CL
bright to CL gray luminescence in all three veinlets. (D, H, sample D211-19): Quartz stockwork vein from the Au-Cu cen-
ter, with a thin K-feldspar selvage in pervasive biotite-altered QMP; vein interior shown in images contains bornite, digen-
ite, and chalcopyrite; cathodoluminescence image reveals complex quartz textures, including euhedral growth zoning in early
quartz generation Q1, followed by fracturing and dissolution surfaces; fracture-filling later generation of quartz Q2 over-
grows earlier zoning and shown faint CL dull growth bands (not visible due to low contrast); sulfides are entirely enclosed in
this later quartz Q2. (I-K, sample D411-2646.5): Pyrite-quartz-bornite-chalcopyrite vein with centimeter-wide sericitic sel-
vages, cutting earlier stockwork veins with potassic alteration in the Au-Cu center; CL petrography reveals three quartz gen-
erations: 1st generation quartz is CL gray to CL dark luminescent and exhibits multiple irregular zoning; later quartz with no
or only weak internal structures, with relatively bright (2nd generation) or dark (3rd generation) luminescence; mapping of
sulfide inclusions reveals that pyrite is restricted to the first generation of quartz, 2nd generation quartz contains bornite and
chalcopyrite inclusions but no inclusions of pyrite; 3rd quartz generation is generally free of any sulfide inclusions.
0361-0128/98/000/000-00 $6.00 100

0361-0128/98/000/000-00 $6.00 101

rock. Cu-Fe sulfides thus precipitated during or after Q1 dis- The spatial distribution of inclusion types varies vertically
solution and before or during recementation of veins and wall and horizontally with reference to the regions defined in Fig-
rocks by Q2. Bright, luminescent Q1 quartz contains signifi- ure 2. Small thumbnail symbols in Figure 7 show the qual-
cantly higher concentrations of trace elements (up to a few itative distribution of intermediate density, vapor, and brine
hundred ppm Ti, Al; tens of ppm Na, K) compared with dull, inclusions in quartz stockwork veins within the representative
luminescent Q2 quartz (Landtwing and Pettke, 2005). In con- cross section 1400E (Redmond et al., 2004; Figs. 3, 7). In-
trast to many other porphyry copper deposits, anhydrite (or termediate-density inclusions predominate in the deep center
gypsum) is absent at Bingham Canyon, and hydrothermal and the deep periphery below the peripheral Cu ore zones.
magnetite is uncommon in the mine, except in some deep Intermediate-density inclusions are by far the dominant in-
drill holes and in minor quantity associated with distal propy- clusion type below ~800 m a.s.l. and occur together with
litic alteration of monzonite. brine and vapor inclusions in an intervening zone, up to an el-
Quartz-molybdenite veins: These veins commonly contain evation of ~1,100 m but still below the Au-Curich center of
open spaces lined by euhedral quartz crystals, which over- the orebody, where they give way to vapor and brine inclu-
grow a more massive granular zone along the vein walls show- sions only. Throughout the Au-Curich center and the upper
ing similar texture and internal growth structures as the stock- parts of the Cu-rich periphery, vapor and brine inclusions
work veins. In contrast to the Cu-Fe sulfides, molybdenite predominate in close association with each other. Occasion-
commonly precipitated before or together with a later generally, vapor and brine inclusions were found to occur on single
ation of bright-luminescent quartz: as symmetric lines of eu- healed microcracks (boiling trails; Audtat et al., 1998; Hein-
hedral flakes growing immediately adjacent to the vein walls, rich et al., 1999), demonstrating that the two fluid types co-
as bands inside the granular quartz, or as free-standing blades existed as immiscible vapor and hypersaline liquid at trapping
overgrowing the euhedral vug crystals. Minor chalcopyrite conditions (see digital supplement). In the periphery of the
(but never bornite) crystallized late in the molybdenite veins, system, intermediate-density inclusions extend upward to
mostly as euhedral crystals in the central vugs. shallower depth, compared to the deep barren center of the
Quartz-pyrite veins with sericitic alteration: These veins system (e.g., sample D169-582). This is indicated by the up-
typically contain only minor amounts of a last generation of ward-concave lines delimiting the small inclusion symbols in
quartz, which occurs as small euhedral crystals lining the vein Figure 7 (cf. Redmond et al., 2004). Distribution boundaries
walls. This quartz is generally dull or nonluminescent and was of fluid inclusion types cross the downward-concave copper
overgrown first by euhedral pyrite and finally by calcite filling grade contours delimiting high-grade copper ore, which ex-
the remaining spaces. Some veins with sericite alteration tends to greater depth toward the periphery of the system
halos have complex transitional characteristics, as exemplified (Fig. 7).
by sample D411-2646.5. Backscattered electron (BSE) and Inclusion distribution in other vein types was studied less
SEM-CL mapping of quartz generations and sulfide inclu- systematically. EDM veins commonly do not have much
sions in this sample (Fig. 5I-K) indicates early CL dull quartz open-space filling quartz, except due to later reopening, and
+ pyrite (generally characteristic of simpler quartz-pyrite the contained brine and vapor inclusions are therefore of un-
veins with sericitic alteration) overgrown by a more lumines- certain association. Quartz-molybdenite veins have essentially
cent second quartz generation with chalcopyrite + bornite in- similar inclusions as the quartz stockwork veins associated
clusions, followed by a third generation of CL dark quartz with Cu Au mineralization. A molybdenite vein cutting a
that lines former vugs filled with chalcopyrite + bornite + late QLP dike in the deep barren center contains intermedi-
pyrite calcite. ate-density inclusions, similar to the earlier formed barren
stockwork veins in EM and QMP in the same region (Table
Distribution of inclusion types and phase proportions 1). Aqueous (type A) fluid inclusions with moderate to small
Inclusion types are classified according to phases seen at bubble size predominate in quartz-bearing pyrite veins asso-
room temperature (Fig. 6). Two-phase inclusions are subdi- ciated with postore sericitic alteration (quartz-pyrite veins
vided into intermediate-density inclusions (type ID; Fig. with sericitic selvages).
6A) containing comparable proportions of liquid and bubble
at room temperature, vapor inclusions (type V; Fig. 6B, D, Phase proportions within two- and three-phase inclusions
F) containing a bubble that occupies the greater part of the trapped in quartz stockwork veins were measured by area
inclusions, and aqueous liquid inclusions (type A; Fig 6H) counting in five representative veins (Fig. 4). Five triangle in-
containing mostly aqueous solution and a smaller vapor bub- sets in Figure 7 illustrate a systematic variation in the density
ble. Many intermediate-density and vapor inclusions con- of two-phase fluid inclusions and in the salinity of brines, con-
tain a small opaque daughter crystal of constant relative size. firming the qualitative observations from other samples (com-
Most type A inclusions are free of true daughter crystals but pare Fig. 6D, E with F, G). Liquid/vapor ratios in the two-
some contain sericite grains that are interpreted to be the phase inclusions, plotting along the upper right edge of each
product of accidental entrapment from ajdacent, sericite-al- triangle, decrease with elevation along the center of the sys-
tered wall rocks. Brine inclusions (type B; Fig. 6C, E, G) al- tem. The trend corresponds to an approximate decrease in
ways contain a vapor bubble, a saline aqueous liquid, and average density from 0.63 g cm3 in intermediate-density flu-
halite. Polyphase brine inclusions contain additional daugh- ids from the deep center, to 0.26 g cm3 for vapor inclusions
ter crystals, such as sylvite, plus transparent or opaque crys- from below the orebody, to 0.19 g cm3 in the Au-Cu center
tals, most notably small red hematite flakes and triangular (using the highest available sample location from a historic
chalcopyrite. core above the present mine surface). On the deep periphery
0361-0128/98/000/000-00 $6.00 102

FIG. 6. Four main types of fluid inclusions measured at Bingham Canyon: Intermediate-density inclusions with similar
proportions of liquid + vapor and a small opaque daughter crystal (A); vapor inclusions with similar appearance but signifi-
cantly greater vapor bubble (B, D, F); brine inclusions with halite + liquid + vapor additional daughter crystals including
sylvite, hematite, and other small transparent and opaque crystals (C, E, G); aqueous inclusions with predominantly liquid
water and smaller vapor bubbles ranging from 20 to 40 vol percent (H). Bottom row of small photos D-G depict typical ap-
pearance of vapor and brine inclusions in the Au-Curich center of the deposit (D, E: thin rim of liquid in vapor inclusions,
high proportion of salts in brine inclusions) compared with those from the Cu-rich periphery (F, G: higher liquid/bubble ratio
in both inclusion types, smaller salt crystal in brine inclusions). Sample numbers: (A) D420-4084; (B, H) D211-19; (C) 208-
37; (D, E) D123-224; (F, G) D188-1277; (H) D411-2646.5.
(at the base of the Cu-rich part of the orebody), two-phase inclusions in the center contain abundant and large salt crys-
fluids are indistinguishable from those in the deep center, but tals, which occupy about the same area in microscope view as
they maintain a greater liquid fraction up to shallower depth. the liquid phase (average h/l ~0.9/1 where h denotes halite +
Phase volume estimates indicate a less pronounced decrease sylvite), whereas brines in the periphery have notably smaller
in average density from ~0.58 g cm3 in intermediate-density halite crystals, usually no sylvite, and a significantly greater
fluids in the deep periphery, to a dense vapor or intermedi- proportion of aqueous liquid (h/l ~0.3/1). Recalculating the
ate-density fluid with ~0.46 g cm3 occurring with brine in the area measurements into approximate volume fractions of salt
Cu-rich periphery near the outer margin of the orebody at crystal(s) and halite-saturated liquid indicates an average total
the elevation of the present open pit. Brine inclusions are salinity of 51 to 54 wt percent NaCl equiv in the Au-Cu cen-
three times less frequent than two-phase inclusions in this ter, increasing from the lower to the higher sample, compared
sample. to only 40 wt percent NaCl equiv in the Cu-rich periphery at
The type B inclusions show an even more conspicuous dif- similar elevation. The average bubble size in the type B in-
ference between center and periphery (Fig. 6E vs. G). Brine clusions, v/(v + l + h), is slightly higher in the Au-Cu center
0361-0128/98/000/000-00 $6.00 103

NW v SE
4000 N
2000 N
0000 N
4000 S
6000 S
2000 S
asl. A B
A57
2200 m
y
rap h
og
A51
top
g v
D74
Pi inin
to
D169
m
ut l
D123
e-
in Pr
D188
20
e
06 6000 ft
D195
s D123-224 Au-Cu center upper part l
D109
D404 D211
D420
D430
1600 m Sedimentary
v host rock
s D188-1277 Cu-rich periphery l

1200 m 4000 ft
First occurrence
of abundant
hypersaline liquid
and vapor-rich
inclusions
v
s D211-221 Au-Cu center, lower part l Sedimentary
host rock
2000 ft
v
Predominance of
Quartz intermediate-density
400 m inclusions
Monzonite
Porphyry
Equigranular
Monzonite
s D420-5192 Deep periphery l
000 m 0000 ft
Statistical analysis of phase proportions Ore grades
Three-phase inclusions v+l+s ~0.15 ppm Au
s A51-4788 Deep center l Single estimates ~0.30 ppm Au
Fluid inclusion types Average ~1.00 ppm Au
Vapor inclusions Two-phase inclusions v+l >0.15 wt.-% Cu
Single estimates >0.35 wt.-% Cu
Brine inclusions Average >0.70 wt.-% Cu
50% envelopes around average
Intermediate-density 100 m
inclusions Lower limit of abundant brine coexisting with vapor
FIG. 7. Distribution of fluid inclusion types and liquid / vapor / halite phase proportions in barren and Cu Au mineral-
ized quartz stockwork veins, plotted on the northwest-southeast geologic cross section of Figure 3. Distribution of dominant
inclusion types is indicated by two thick lines and thumbnail symbols; phase proportions of vapor bubble (v), halite + other
salt crystals (s), and aqueous liquid (l), measured by area percent, are plotted in triangular diagrams for five samples. Large
black symbols and line contours in the triangular plots denote averages and 50 percentile envelopes of up to 60 inclusions of
each type measured per sample.
(0.26 by area fraction) compared to the periphery at similar than vapor-bubble disappearance were observed at Bingham
paleoelevation (0.21), in line with the density difference esti- Canyon, which in other deposits is clear evidence for posten-
mated for the vapor inclusions between center and periphery. trapment modification of inclusions (e.g., Audtat and Gn-
ther 1999; Klemm et al., 2007, 2008).
Microthermometric data Intermediate-density (type ID) inclusions show reasonably
Microthermometric data were combined from several consistent phase transition temperatures within assemblages,
stages of investigation and are individually documented in the but the measurements are difficult to interpret. Minor CO2 is
theses of Redmond (2002), Landtwing (2004), and Furrer ubiquitous, as indicated by melting events close to 56.6
(2006). Clearly definable petrographic assemblages and good 0.5C and consistent clathrate melting between 1 and 8C.
consistency within most inclusion assemblages indicate good This does not uniquely determine inclusion salinity, because
preservation of composition and bulk densities. No boiling the CO2 content is too small to saturate a well-visible liquid
trails with halite dissolution temperature significantly higher CO2 phase. Figure 8 (based on Diamond, 1992) summarizes
0361-0128/98/000/000-00 $6.00 104

10 1 wt percent NaCl equiv (Furrer, 2006). Intermediate-den-

Moly vein A57-5910
0 sity inclusions in one assemblage in the deep barren center of
8 2 A57-5889.5 the deposit indicate a relatively high salinity of ~13 wt per-
Vapor 4
D169-582
cent NaCl equiv based on a low temperature of last clathrate
6 other D74-1719 6 melting in the presence of liquid CO2 (Redmond, 2002; sam-
Tm,clathrate ( C)
? 8 Center ple A57-6063), but numerous other intermediate-density as-

4 Intermediate-density semblages in the same region have higher clathrate melting
D430-3657 Periphery
2
(58.5C). Given the likelihood that liquid CO2 was over-
D430-3749 A57-6063
12 looked in these commonly small inclusions, this would indi-
0 14 cate salinities below about 5 2 wt percent NaCl equiv (Fig.
8). Intermediate-density inclusions beneath the central Au-
-2 16 % NaCleq in aqueous phase Cu orebody homogenize by meniscus fading or by bubble ex-
pansion to the vapor phase at temperatures varying from
-4 ~365 to 429 14C (or even higher in sample A57-5889.5).
-20 -16 12 -8 -4 0 4 8 Such homogenization behavior above 380C is only possible
Th,CO2(L+V >V) (C) if the inclusions contain significant salt as well as CO2 (Gehrig,
1980). Intermediate-density inclusion in several samples from
FIG. 8. Diagram based on figure 5 of Diamond (1992), showing isolines of the deep base of the peripheral Cu zone (5.87.2 wt % NaCl
fluid salinity as a function of the last melting of clathrate (Tm(clathrate) vertical
axis) against the temperature of homogenization of the CO2 phase from L +
equiv) homogenize to the liquid phase within a narrow range
V to V (Th(CO2); dew-point homogenization of the carbonic phase). Limited of 361 4 to 384 11C (Furrer, 2006; Fig. 9). This indi-
constraints on salinities of intermediate-density (type ID) and vapor (V) in- cates that the deep single-phase fluids in the periphery have
clusions at Bingham Canyon are plotted as boxes and discussed in the text. slightly higher density but lower temperature (at similar
depth and composition), compared with the single-phase flu-
ids in the barren core below the center of the deposit.
the few measurements constraining the salinity of two-phase The molybdenite-bearing vein cutting a late QLP dike in
(L + V) inclusions of intermediate or vaporlike density, based the deep center (sample A57-5910) contains intermediate-
on observations of CO2 and clathrate melting. The clearest density inclusions homogenizing by meniscus fading or to the
salinity measurements were obtained from two assemblages vapor at 390 30C. These inclusions contain liquid CO2
of intermediate-density inclusions in the Cu-rich periphery of and clathrate with measurable temperatures, constraining
the deposit, where the homogenization of the CO2 phase and their salinity to 2.5 2.5 wt percent NaCl equiv (Fig. 8).
the last melting of clathrate were both unambiguously and re- Vapor inclusions (type V) are the predominant inclusion
producibly measured, indicating salinities of 5.8 1.4 and 7.2 type in all stockwork veins of the Bingham Canyon orebody.
600
bar
700 bar
600
bar
500
Trapping
500 ID fluids ar
400 b
Cu-rich periphery 300 bar

T (C)
deep center
400 Vapor ID fluids
Au-Cu center deep 200 bar
periphery
100 bar Brine

300
Au-Cu center of orebody
Typical 1 Cu-rich periphery of orebody Au-Cu center, boiling pair
uncertainty Late brine, deep periphery Late brines, deep center
in assemblage
EDM vein, Au-Cu center Below orebody, center
200
0 10 20 30 40 50 60
Apparent Salinity (NaCl wt% eq.)
FIG. 9. Summary of microthermometric data, showing salinity in wt percent NaCl equiv vs. temperature of final homog-
enization; salinities determined from halite dissolution (Bodnar and Vityk, 1994; brine inclusions, type B) and rare observa-
tions of clathrate melting constrained by liquid CO2 in ID and V type inclusions (Fig. 8); each point symbol represents one
assemblage, usually averaging several individual inclusion measurements; fine background lines are isobars of the limiting
two-phase (brine + vapor) surface and the halite saturation limit of the NaCl-H2O model system (after Driesner and Hein-
rich, 2007).
0361-0128/98/000/000-00 $6.00 105

In contrast to the generally small intermediate-density inclu- Two-phase aqueous (type A) inclusions are minor but wide-
sions, the vapor inclusions are commonly large with a big and spread as secondary inclusions in all veins, demonstrably
easily observable bubble. A single assemblage contained suf- postdating Q2 quartz in stockwork vein samples studied by
ficient CO2 to allow homogenization as well as clathrate tem- CL. They are the predominant and locally primary inclusions
peratures to be measured, yielding a poorly defined low salin- associated with the dull-luminescent third quartz of late
ity of 0.9 1 wt percent NaCl equiv. Many other vapor quartz-pyrite veins with sericitic alteration (QSP veins: Fig.
assemblages show clathrate melting between 4.4 0.4 and 5I). Phase proportions and reconnaissance measurements in-
7.4 0.3C, without evidence of liquid CO2 despite good vis- dicate low salinity and homogenization temperatures of
ibility and careful observation, which indicates that they also ~320C or lower.
have low salinities (less than ~4 wt % NaCl equiv; Fig. 8).
Final homogenization occurs by bubble expansion but was Elemental composition of fluids
not systematically measured, because only minimum temper- Cation ratios relative to Na, determined by LA-ICPMS mi-
atures are observable. croanalysis, are plotted in Figure 10 (two-phase inclusions of
Brine inclusion (type B) data are plotted in Figure 9 as tem- types ID, V and A) and Figure 11 (brine inclusions). Each
perature of final homogenization versus apparent salinity in wt point represents one inclusion assemblage, with error bars
percent NaCl equiv, whereby each point symbol represents the denoting 1 variation of individual inclusions in the assem-
average from an assemblage of 1 to 12 individual inclusion blage. Data are sorted first according to sampling region, then
measurements. Uncertainty within one assemblage is typically by sample and veining stage. Within one sample, data are
2 wt percent NaCl equiv and 8C in Th. The total range in sorted arbitrarily by decreasing Cu/Na ratio. Estimates of
salinity and homogenization temperature varies from 32 to 57 fluid compositions in different regions of the deposit are
wt percent NaCl equiv and from 260 to 550C, respectively. given in Tables 2 and 3, based on a combination of LA-
This is much greater than within-assemblage uncertainty, ICPMS measurements of metal ratios, microthermometric
demonstrating true geologic variations. Most of the inclusion data, and mass-balance constraints based on the interpreted
assemblages from the central Au-Cu orebody (Au-Cu center, fluid evolution (below).
Fig. 2) homogenize by vapor disappearance into the liquid, but The concentration ratios among most major and trace ele-
there are also many assemblages, especially at the lower tem- ments are essentially constant throughout all generations of
perature end of the range, that finally homogenize by halite fluid inclusions. This is particularly prominent in the brine in-
dissolution after bubble disappearance. Unambiguous boiling clusions where the LA-ICPMS signals are strong and repro-
trails with coexisting brine and vapor inclusions commonly ducibility within one assemblage is best. Also among two-
show almost simultaneous disappearance of halite and bubble phase (types ID, V and A) inclusions, there are no systematic
to a homogeneous liquid, suggesting transient halite saturation differences in concentration ratios among Na, K, Fe, Mn, Zn,
in the central Au-Cu orebody. This would constrain the trap- Pb, Rb, Cs, and additional elements analyzed close to their
ping conditions to a low pressure of ~150 bars at ~380C (Fig. detection limit (Ca, Ba, Sr, and possibly As, W).
9, lowermost large squares). The highest temperatures are ob- By contrast, the main ore-forming metals Cu, Mo, and
served in early brine assemblages trapped in presulfide Q1 probably also Au show significant variations between assem-
quartz of the high-grade Au-Curich center, but we have no blages in each sample and also from sample to sample be-
petrographic record of whether they coexisted with associated tween different regions. Copper is a major component be-
vapor inclusions (Redmond, 2002). Brine salinities in the cen- sides Na, K, and Fe in type ID, B, and V fluids and is even the
ter cluster at 45 5 wt percent NaCl equiv, whereas brines in dominant cation element in some vapor inclusion assem-
the Cu-rich periphery are of lower salinity on average (35 4 blages (Cu/Na >1). Cu/Na varies over three orders of magni-
wt % NaCl equiv) but homogenize at similar temperatures tude, even in single samples. This variation is geologically
(340420C) as the majority of assemblages in the Au- real, being much greater than the variability within one as-
Curich center. These lateral salinity differences are consistent semblage (although the within-assemblage variation in Cu/Na
with the petrographic observations of higher salt proportions in is generally greater than between other cations; note error
the Au-Curich center, compared to the periphery (Fig. 7). bars in Figs. 10 and 11). Copper concentration in brine inclu-
Rare boiling-trail assemblages (too small for quantitative mea- sions correlates with the temperature of final homogenization
surement) indicate that some fluids in the Cu-rich peripheral for those inclusion assemblages in sample D211-19 that could
orebody were also trapped in the two-phase field of coexisting be associated with the early quartz generation Q1 (Landtwing
liquid + vapor but well below halite saturation (Fig. 9). et al., 2005; see discussion below). Molybdenum concentra-
A thin EDM veinlet within QMP contains brines with the tions are much lower and not as well defined, but Mo/Na and
highest of all observed salinities but without clearly coeval also the ore metal ratio Mo/Cu seem to reach their highest
vapor inclusions. Final homogenization by halite dissolution values in some of the fluid inclusion assemblages from quartz-
as high as 450C indicates salinities up to 58 wt percent NaCl molybdenite veins. Gold is invariably close to the detection
equiv (Fig. 9), but larger than usual variation within assem- limit and only detected in some inclusions in each assem-
blages may indicate postentrapment modification of these blage. This creates an ambiguity in averaging of results, as
petrographically early inclusions. the clearly detected gold signals in each assemblage are
Brine inclusions in quartz-molybdenite veins were not stud- likely to be at the upper end of the true analytical variation
ied in detail and are omitted from Figure 9. Microthermo- within the assemblage. A statistical tendency to overrepre-
metric data overlap the range of inclusion properties in quartz sent the highest values is also indicated by the individually
stockwork veins in the central region (digital supplement). calculated limits of detection, expressed as maximum Au/Na
0361-0128/98/000/000-00 $6.00 106

Quartz stockwork veins Quartz-molybdenite QSP vein
1 2 3 4 5 4 1 4 4
A57-5889.5 D430-3749 D430 D188-4491 D109-1039 D195-1032 D211-19 5190 D74-1719 D169-852 D211 D195- A57-5910 4990 D211 D411
X/Na -3657 -1600 -19 1386 -2050 -176 -2646.5
1 Na
K
Fe
0.1 Mn
0361-0128/98/000/000-00 $6.00
0.01 Intermediate-density inclusions
Vapor-rich inclusions
0.001 Petrographically late aqueous liquid-rich inclusions
X/Na
10
1
Cu
0.1 Zn
Pb
0.01
Mo
0.001
Ag
Au
107
0.0001
0.00001
X/Na
0.1
As
Ba
0.01
Rb
Sr
0.001 W
Cs
0.0001
0.00001
BINGHAM CANYON: ZONED Cu-Au ORE DEPOSITION BY MAGMATIC VAPOR EXPANSION
1 Deep centre 2 Deep periphery 3 Below orebody 4 Au-Cu centre 5 Cu-rich periphery
FIG. 10. Element ratios X/Na for two-phase fluid inclusion assemblages (types ID, V, and A) in stockwork veins from different regions of the system (Fig. 2, Table 1;
digital supplement), Also shown are type V inclusions from two molybdenite veins and aqueous liquid inclusions (type A) in a late quartz-pyrite (QSP) vein. Each sym-
bol represents one assemblage, averaged from 1 to 15 values measured above the detection limit, which is calculated for each element in each inclusion (Heinrich et al.,
2003). Error bars indicate 1 where two or more values were determined. Within one sample, assemblages are arbitrarily sorted by decreasing Cu/Na.
107
Quartz stockwork veins Quartz-molybdenite veins 108
1 2 3 4 5 1 3 4
A57-5889.5 D404-1632 D109-1039 D195-888 D195 D211-19 D74-1719 D74-1587 A57 D195-1386, D195-1386A 4990 D211 wt%
C -1032 -5910 -2050 -176
5190
NaCl(eq)
D188
D430
-4491
-1600
-3749
500 60
55
50
400 TL
TNaCl
40
300
35
0361-0128/98/000/000-00 $6.00
200
X/Na
1 Na
K
Fe
0.1
Ca
Mn
0.01
X/Na
1
0.1 Cu
Zn
0.01 Pb
Mo
108
0.001 Ag
Au
0.0001
LANDTWING ET AL.
0.00001
0.000001
X/Na
0.01
Rb
Ba
Sr
0.001
W
As
Cs
0.0001
1 Deep centre 2 Deep periphery 3 Below orebody 4 Au-Cu centre 5 Cu-rich periphery
FIG. 11. Summary of microthermometric data (top panel) and element ratios X/Na for brine inclusion assemblages (type B) in an EDM vein, in stockwork veins from
different regions of the ore system (Fig. 2, Table 1; digital supplement), and in molybdenite veins from the center of the orebody, sorted by decreasing Cu/Na within
each sample. Error bars indicate 1 of several inclusions in one assemblage. Element ratios are plotted for easier comparison with the two-phase inclusions shown in
Figure 10 but relate more closely to absolute concentrations, as total salinities of the brines vary within less than a factor of two.
in each inclusion, which are commonly smaller for some in- deposit. Deposit-scale variation of Au/Cu ratio in the ore is
clusions than the calculated average of Au/Na ratios of de- explained by slightly different evolution paths of a common
tected gold signals. However, gold is unquestionably present input fluid ascending through a broad column of fractured
as an inclusion component at a concentration ratio of Au/Na rock extending well beyond the QMP. The single-phase input
~1051, but variations within this uncertainty range may not fluid was initially hot and under lithostatic pressure, with high
be geologically significant. Au and Cu contents controlled by a deep igneous source. Por-
Comparing brine and vapor inclusions in samples contain- tions of this fluid ascending along the central zone of highest
ing both fluid inclusion types, the vapors are on average en- permeability expanded above the transition from lithostatic to
riched in Cu/Na compared with the brines. Figure 12 shows hydrostatic fluid pressure, resulting in wholesale precipitation
the data from a single boiling trail, i.e., from strictly coeval of both metals in the central high-grade Au-Cu orebody. Ore
brine and vapor inclusions trapped on a single healed mi- deposition occurred at low pressure, dominantly from the ex-
crofracture in Q1 quartz from the central Au-Cu orebody panding vapor and to a lesser degree from minor condensing
(sample D211-19; Table 1, digital supplement). Both fluids brine. Portions of similar fluid ascending in the periphery of
contain high Cu/Na, but the vapor inclusions are enriched in the vein network cooled at slightly higher pressure and un-
Cu relative to Na and all other metals. The majority of indi- derwent less extreme phase separation, resulting in selective
vidual vapor inclusions on this trail contain Cu as the most copper precipitation in the deeper peripheral ore zones. The
abundant metal. Au also tends to fractionate in favor of the following sections discuss key aspects of the proposed process
vapor phase, relative to Na, but only a few analyses returned and place them into the broader context of porphyry copper
detectable signals, leading to large uncertainty. Additional gold ore formation.
boiling trail data are reported in Seo et al. (2009), showing
high Cu, Au, as well as S concentrations in both fluid phases One homogeneous deep input fluid
and a correlated enrichment of all three ore-forming ele- The input fluids to the main stage of Cu Au mineraliza-
ments relative to Na in the vapor. tion were sampled ~3 km below the land surface at the time
of mineralization (Fig. 2; Deino and Keith, 1997; Waite et al.,
Interpretation and Discussion 1997; Gruen et al., 2010, fig. 1C). The intermediate-density
The fluid inclusion data together with geologic observations (type ID) inclusions trapped a single-phase fluid at near-litho-
documented in Redmond et al. (2001), Redmond (2002), static pressure, estimated to be around 820 bars based on the
Redmond and Einaudi (2010), and Gruen et al. (2010) can ex- stratigraphic overburden. Homogenization to vapor or by
plain the origin of the zoned ore shell of the Bingham Canyon meniscus fading at 400 30C and an apparent salinity of ~7
wt percent NaCl equiv are inconsistent with the binary NaCl-
H2O model system in which a fluid of such high salinity would
10
homogenize into the liquid phase (Fig. 9; Driesner and Hein-
Cu rich, 2007). Estimating the actual trapping temperature of the
1 deep input fluid requires a pressure correction, which is dif-
Fe Na ficult to calculate given this compositional complexity. The
K
isochore of a 7 wt percent NaCl liquid homogenizing at 400
0.1 30C provides a likely minimum correction, as the iso-
Zn chores of a CO2-bearing vapor-homogenizing fluid would
Pb Mn
(M/Na)Vapor
have a somewhat lower dP/dT slope. Using the SoWat iso-

0.01
chore equations of Driesner and Heinrich (2007) for the bi-
nary model system yields a fluid density of 0.59 0.07 g cm3,
10-2
W in good agreement with the graphically determined average
Au
Bi
of 0.63 in sample A51-4788 (Fig. 7). Following this isochore
to ~820 bars yields a temperature correction of about 90C.
-3
10 Added to the range of measured homogenization tempera-
tures, this correction defines the minimum trapping condi-
tions of the deep input fluid, at 450 to 520C in the single-
10-4
phase stability region (Fig. 9).
The elemental composition of the input ore fluid is sum-
10-5 marized in Figure 13, comparing the ranges between inter-
10-5 10-4 10-3 10-2 0.01 0.1 1 10
mediate-density inclusion assemblages from the barren deep
center below the Au-Cu orebody with those from the deep
(M/Na)Brine periphery at the base of the Cu-rich peripheral ore zone.
FIG. 12. Element ratios X/Na in vapor plotted against the corresponding Even excluding some outliers, the deep single-phase fluids
concentration ratio in brines from a high-temperature (426C) boiling trail in from both regions have overlapping composition in terms of
Q1 quartz from the stockwork vein D211-19, trapped on a single healed conservative elements like K, Pb, or Cs, as well as the ore-
crack and representing a two-phase fluid at the onset of bornite-chalcopyrite forming elements Cu, Mo, Ag, and Au. Copper in deep inter-
precipitation. Although Cu concentrations are variable and Au was only de-
tected in a few of the inclusions, the vapor is enriched in these ore metals rel- mediate-density inclusion assemblages varies significantly in
ative to the brine and contains Cu as the predominant cation (Cu/Na > ~1 totally Cu-Fe sulfide-free samples, and the Cu/Na variation
corresponding to a concentration of ~1 wt % Cu in the vapor). even within one assemblage is commonly greater than other
0361-0128/98/000/000-00 $6.00 109

~100 to 200 bars. Brine and vapor inclusions in the Cu-rich

Cu periphery of the deposit are more similar in density (Fig. 7)
1 K Fe
and salinity (Figs. 8, 9). Thus, brine and vapor in the periph-
Element / Na
ery were trapped at conditions where the two-phase region

0.1 was narrower, implying higher pressures of ~200 to 300 bars
Pb at similar temperatures of 350 to 420C. These spatial dif-
Mo
10-2 Ag ferences in fluid pressure are likely to be real, even though
the absolute pressure-temperature determinations are not
Au well defined. Individual boiling trails potentially give more
10-3 Cs precise measurements of pressure-temperature conditions,
Deep barren center but minor unquantified CO2 and salts other than NaCl also
10-4 lead to uncertainties up to 100 to 200 bars (e.g., Liebscher,
Deep periphery
2007).
Average all ID assemblages
The spatial variation of fluid density and composition in
10-5
the Bingham Canyon system indicate upflow and phase sep-
FIG. 13. Ranges of element ratios in intermediate-density (type ID) in- aration along a steep decompression gradient below the Au-
clusion assemblages, showing that deep input fluids in the center and at the Curich center of the orebody. By contrast, originally similar
periphery of the ore system are indistinguishable, regarding ore metals (Cu, fluids ascending along the Cu-rich periphery of the orebody
Mo, Ag, Au) as well as conservative elements in the porphyry orebody re- followed a more gentle decompression and fluid-separation
flecting unmodified source characteristics (K, Fe, Pb, Cs). Small diamonds
indicate averages of all assemblages used as common input fluid for the mass-
path. Mass-balance calculations shown in Tables 2, 3, and
balance calculation in Tables 2 and 3 (first columns), except for Cu where the Figure 14 confirm that the interpreted phase separation of a
average of three consistently high assemblages used to account for incipient deep single-phase fluid into brine and vapor is consistent
sulfide precipitation, and excluding very high Au outliers (possibly due to with all analytical constraints, not only regarding total salinity
trapped microparticles). but also for all analyzed element ratios. The two tables list
best-fit results from a spreadsheet model based on mass con-
servation for all components, using assumptions that are de-
major cation ratios. This variation could indicate variation in tailed in footnotes to Table 2 (Excel spreadsheet in digital
fluid source composition or incipient phase separation, or be supplement). We take as given the average apparent salinity
due to some postentrapment diffusive reequilibration of of the brines (in wt % NaCl equiv), the empirical approxima-
cation ratios in these high-temperature inclusions (Li et al., tion for the effect of other salts upon microthermometric be-
2009; Zajacz et al., 2009). Gold was clearly detected in the in- havior (Heinrich et al., 2003), and the average element ratios
clusions from the deep periphery, showing that the peripheral
input fluids were not inherently Au poor compared with those
deep below the Au-Cu center of the deposit. Comparing the 10
Vapor compositions
M / Na (observeed)
average Au content of all deep intermediate-densitytype

center
fluid inclusions with the upper end of the range of Cu values Periphery Na
yields an Au/Cu ratio of ~104. This value is close to the bulk 1 Fe
K Cu
ore Au/Cu ratio of 0.8104 contained within the Au-Cu cen-
ter of the orebody (inside the >0.6 g/t Au contour; Gruen et
al., 2010). 0.1 Mn
Pb
Phase separation by condensation of minor brine Ba
B
Zn
The mapped transition from intermediate-density inclu- 0.01 W As

sions in deep quartz stockwork veins to the ubiquitous associ- Rb
Sr Mo
ation of brine and vapor inclusions higher up (Fig. 7) is inter-
Ag
preted as a phase separation front within an ascending fluid 103 Au
plume. The upward-concave shape of this front, the variable
phase proportions in the inclusions at room temperature (Fig.
Cs
7), and the microthermometric data from associated and lo-
104
cally coexisting vapor and brine inclusions (Fig. 9) can be in-
terpreted by comparison with the two-phase stability surface 104 103 0.01 0.1 1 10
(solvus) of the NaCl-H2O model system (thin isobars shown M / Na (predicted)
in Fig. 9 from Driesner and Heinrich, 2007).
FIG. 14. Ore-forming vapor compositions, comparing analytical observa-
The data indicate different degrees of phase separation be- tion with predictions from mass balance during phase separation in the cen-
tween the center and the periphery of the fluid plume, which tral upflow zone (full symbols) and in the Cu-rich periphery (open symbols,
is very sensitive to pressure in this pressure-temperature re- limited data). Observed element ratios in vapor inclusions from the two re-
gion. In the center, phase separation was most extreme, lead- gions are plotted against the predicted vapor composition (Tables 2, 3) that
would result from subtraction of 10 mass units (X = 0.10) of analyzed brines
ing to highly saline brine coexisting with a vapor of low den- from 100 mass units of single-phase intermediate-density fluid (indistin-
sity and low salinity. The observations indicate a wide gap of guishable composition in the deep barren center and at the base of the pe-
fluid unmixing in the central upflow zone, at pressures of ripheral copper orebodies; see Fig. 13 and text for discussion).
0361-0128/98/000/000-00 $6.00 110

TABLE 2. Best-Estimate Ore Fluid Compositions in the Stockwork Vein Stage in the Central Upflow Zone and the Au-Cu Orebody Prior to Au and Cu-Fe Sulfide Saturation
Compositions are calculated with a spreadsheet to be consistent with a common input fluid based on the average element ratios in intermediate-density fluids measured by LA-ICPMS (column c),
followed by phase separation into brine and vapor; brine composition is based on averaged analyses of inclusion from the deposit center (columns e, f), and the vapor is calculated by mass-balance
(columns h, i) and compared with analyzed element ratios (column l); see footnotes for calculation, Figure 13 for comparative plot, and text for discussion
Intermediate-density fluid Brine Vapor predicted Vapor analyzed

analyzed (avg all ID) analyzed (avg all B below orebody) (100 g ID fluid X 100 g brine) (avg all V)
App Sal % App Sal % App Sal %

NaCl equiv NaCl equiv Fluid mass NaCl equiv Fluid mass
0361-0128/98/000/000-00 $6.00
7.0 45.0 fraction X 3.88 fraction 1-X
Na wt % = 1.97 Na wt % = 11.87 0.10 Na wt % = 0.87 0.90
Concentration Concentration Concentration Uncertainty 1

Com- MMCl/MM El ratio wt % El ratio wt % Component El ratio wt % Component El ratio
ponent M/Na (ICP) (gMCl/100g ID) M/Na (ICP) (gMCl/100g B) fraction M/Na (pred) (gMClx/100gV) fraction M/Na (ICP) +
(a) (b) (c) (d) (e) (f) (g) (h) (i) (k) (l) (m) (n)
NaCl 2.5421 1 4.9963 1 30.165 3.0165 1 2.1998 1.9798 1

KCl 1.9067 0.5500 2.0611 0.7500 16.969 1.6969 0.2453 0.4047 0.3642 0.6300 0.3400 0.3400
FeCl2 2.2696 0.3500 1.5613 0.3800 10.234 1.0234 0.3043 0.5976 0.5379 0.4900 0.1600 0.1600
MnCl2 2.2907 0.0380 0.1711 0.0470 1.2775 0.1278 0.0243 0.0481 0.0433 0.0370 0.0200 0.0200
Cu 1 0.7500 1.4741 0.0630 0.7476 0.0748 1.7967 1.5548 1.3993 1.9400 5.0500 1.9000
ZnCl2 2.0847 0.0290 0.1188 0.0270 0.6679 0.0668 0.0320 0.0578 0.0520 0.0360 0.0210 0.0210
PbCl2 1.3422 0.0166 0.0438 0.0228 0.3631 0.0363 0.0072 0.0083 0.0075 0.0250 0.0200 0.0200
Mo 1 0.0043 0.0085 0.0012 0.0142 0.0014 0.0090 0.0078 0.0070 0.0051 0.0094 0.0050
AgCl 1.3287 0.0025 0.0065 0.0002 0.0032 0.0003 0.0061 0.0069 0.0062 0.0028 0.0056 0.0027
Au 1 0.0001 0.0003 0.0000 0.0002 0.0000 0.0003 0.0003 0.0002 0.0012 0.0029 0.0011
111
Rb 1.4148 0.0034 0.0095 0.0045 0.0755 0.0076 0.0017 0.0021 0.0019 0.0085 0.0068 0.0068
Ba 1.5163 0.0023 0.0069 0.0011 0.0198 0.0020 0.0041 0.0054 0.0049 0.0230 0.0420 0.0220
Sr 1.8092 0.0025 0.0089 0.0010 0.0215 0.0021 0.0048 0.0075 0.0067 0.0041 0.0058 0.0040
W 1 0.0025 0.0049 0.0014 0.0166 0.0017 0.0039 0.0036 0.0033 0.0170 0.0520 0.0165
As 1 0.0025 0.0049 0.0003 0.0036 0.0004 0.0058 0.0051 0.0046 0.0120 0.0110 0.0110
Cs 1.2668 0.0004 0.0010 0.0003 0.0045 0.0005 0.0006 0.0006 0.0005 0.0004 0.0003 0.0003
B 1 0.0091 0.0178 0.0012 0.0142 0.0014 0.0211 0.0183 0.0164 0.0300 0.0290 0.0290

TDS wt % 10.4777 TDS wt % 60.5827 TDS wt % 4.9105
Calculation procedure and assumptions (all data in wt units; absolute concentrations in wt %, i.e., grams per 100 g fluid; see digital supplement)
1. Element ratios M/Na in columns (c) and (e) are assemblage averages directly form LA-ICPMS analysis: (c) = all ID from deep barren center and deep periphery Cu average from three consis-
tently high assemblages (low values in range may indicate incipient sulfide deposition); outlying high Au values (microparticles?) omitted; values in (e) are all brines from stockwork veins below the
Au-Cu orebody, i.e., after phase separation but prior to significant Cu-Au deposition
2. Apparent salinity (wt % NaCl equiv) shown as bold figures near top of columns (d) and (f) are adjusted for best fit within the ranges indicated by microthermometry; values of 7 2 for ID fluid
and 45 3 for brines predict positive concentrations in vapor (column i), and predicted ratios (h) in agreement with analyses (column l)
3. Na wt % (italics, top of d, f) computed from NaCl wt % equiv (apparent salinity, bold) by empirical correction, using conversion factors (b) and analyzed ratios M/Na: Na wt % = NaCl wt %
BINGHAM CANYON: ZONED Cu-Au ORE DEPOSITION BY MAGMATIC VAPOR EXPANSION
equiv/[MNaCl/MNa + 0.5*(M/Na*MMCl/MM)] summed over all major components except Cu (Heinrich et al., 2003)
4. Absolute concentrations (columns d, f) of all salts or elements (column a) are calculated from element ratios (c, e) and the absolute Na wt % concentration
5. Assuming mass fractions of Brine (X) and Vapor (1-X) by splitting 100 g of ID fluid, predict fraction of each component in Vapor (k) by difference: component fraction in Vapor (in grams, col-
umn k) = component concentration wt % in ID fluid (column c) component fraction separated into Brine (column g)
6. Back-calculate (k) into concentrations of each component in Vapor (column i) and predicted element rations of Vapor (column h); TDS = total disolved salts
7. Adjust apparent salinities of ID and B for positive values of predicted element ratios (h) and compare with analyzed element ratios (l = avg all V in center; uncertainties of analyzed vapors +
1, 1 or estimated lower value >0); check whether apparent salinity is consistent with microthemometry (<4 wt % NaCl equiv)
111
112
TABLE 3. Best-Estimate Ore Fluid Compositions in the Stockwork Veins of the Peripheral Cu Ore Shell
0361-0128/98/000/000-00 $6.00
Calculated in same way as Table 2 and based on the same intermediate-density input fluid, but lower salinity brine inclusions as observed in peripheral parts of the system
Intermediate-density fluid Brine Vapor predicted Vapor analyzed

analyzed (avg all ID) analyzed (avg all B in Cu-rich periphery (100 g ID fluid X 100 g brine) (avg all V)
App Sal % App Sal % App Sal %

NaCl equiv NaCl equiv Fluid mass NaCl equiv Fluid mass
7.0 36.0 fraction X 5.34 fraction 1-X
Na wt % = 1.97 Na wt % = 8.60 0.10 Na wt % = 1.23 0.90
Concentration Concentration Concentration Uncertainty 1

Com- MMCl/MM El ratio wt % El ratio wt % Component El ratio wt % Component El ratio
ponent M/Na (ICP) (gMCl/100g ID) M/Na (ICP) (gMCl/100g B) fraction M/Na (pred) (gMClx/100gV) fraction M/Na (ICP) +
(a) (b) (c) (d) (e) (f) (g) (h) (i) (k) (l) (m) (n)
NaCl 2.5421 1 4.9963 1 21.850 2.1850 1 3.1238 2.8115 1

KCl 1.9067 0.5500 2.0611 0.8111 13.293 1.3293 0.3471 0.8132 0.7319 0.6100 0.2400
112
FeCl2 2.2696 0.3500 1.5613 0.6918 13.496 1.3496 0.0844 0.2353 0.2118 0.5700 0.3800 0.3800
MnCl2 2.2907 0.0380 0.1711 0.0329 0.6484 0.0648 0.0422 0.1189 0.1070 n.d.
Cu 1 0.7500 1.4741 0.1740 1.4954 0.1495 1.1977 1.4717 1.3246 0.3100 0.1600 0.1600
LANDTWING ET AL.
ZnCl2 2.0847 0.0290 0.1188 0.0253 0.4524 0.0452 0.0314 0.0804 0.0724 n.d.
PbCl2 1.3422 0.0166 0.0438 0.0266 0.3065 0.0307 0.0089 0.0146 0.0131 0.0210 0.0090 0.0090
Mo 1 0.0043 0.0085 0.0002 0.0021 0.0002 0.0075 0.0092 0.0082 0.0009
AgCl 1.3287 0.0025 0.0065 0.0003 0.0029 0.0003 0.0043 0.0069 0.0062 n.d.
Au 1 0.0001 0.0003 0.0000 0.0002 0.0000 0.0002 0.0003 0.0002 0.0008 0.0003 0.0003
Rb 1.4148 0.0034 0.0095 0.0045 0.0542 0.0054 0.0026 0.0045 0.0040 n.d.
Ba 1.5163 0.0023 0.0069 0.0009 0.0111 0.0011 0.0034 0.0064 0.0057 n.d.
Sr 1.8092 0.0025 0.0089 0.0010 0.0157 0.0016 0.0037 0.0081 0.0073 n.d.
W 1 0.0025 0.0049 0.0008 0.0072 0.0007 0.0035 0.0047 0.0042 n.d.
As 1 0.0025 0.0049 0.0005 0.0045 0.0004 0.0039 0.0050 0.0045 n.d.
Cs 1.2668 0.0004 0.0010 0.0005 0.0052 0.0005 0.0003 0.0005 0.0005 0.0004 0.0002 0.0002
B 1 0.0091 0.0178 0.0027 0.0230 0.0023 0.0141 0.0173 0.0155 n.d.

TDS wt % 10.4955 TDS wt % 51.6680 TDS wt % 5.9207
analyzed in the intermediate-density and the brine inclusions, the liquid side or on the vapor side of the critical curve (e.g.,
whereby the maximum Cu/Na of the brines was taken to rep- Liebscher and Heinrich, 2007). A homogeneous fluid with
resent the phase-separated fluid prior to Cu-Fe sulfide pre- initially 7 wt percent salinity, ascending from 800 bars at any
cipitation. We then varied the apparent salinity of the inter- temperature greater than ~450C, will first intersect the two-
mediate-density fluids within their possible limits and the phase curve on the vapor side of the critical curve (Fig. 9).
unknown mass fractions of vapor/brine formed by separation Therefore, phase separation at Bingham Canyon occurred
of a given amount of intermediate-density fluid. For selected not by boiling (i.e., formation of vapor bubbles from a liquid)
values for these two parameters, the resulting salinity and all but rather by condensation of droplets of hypersaline brine
metal concentrations of the vapor are predicted by the model from the dominant vapor.
and can be compared to the plausible salinity range and ana-
lyzed element ratios in the vapor inclusions (Fig. 14). Sulfide precipitation during quartz dissolution
The mass balance on phase separation was calculated inde- Cathodoluminescence textures of ore samples consistently
pendently for the interpreted central upflow zone (Table 2) show that chalcopyrite and bornite precipitated into sec-
and for the periphery of the system (Table 3), taking into ac- ondary pore space, created by dissolution of the main vein-
count the observed difference in average brine salinity of 45 forming quartz generation Q1 (Redmond et al., 2004). Min-
and 36 wt percent NaCl equiv, respectively. Results showed eral textures in combination with fluid inclusion data,
that only a limited range of input salinity and vapor/brine showing correlated decreases in temperature and Cu content
ratio satisfy all constraints, without predicting negative con- of secondary fluid inclusion assemblages in Q1, indicate that
centrations of some elements or salinities outside the range Cu-Fe sulfide precipitation occurred during or immediately
permitted by microthermometric data. The independently after Q1 dissolution. This was followed by recementation of
predicted vapor compositions are therefore well constrained, the vein with a second quartz generation, Q2, during or after
and the results agree with the analyzed inclusion composi- Cu-Fe sulfide precipitation (Landtwing et al. 2005).
tions (Fig. 14). Figure 15 links these textural observations with limited ex-
Both in the center and in the periphery, the best fit is perimental solubility constraints. In pure water, quartz solu-
achieved assuming a phase ratio of vapor/brine of approxi- bility decreases monotonically with falling temperature at
mately 9/1. Including the modeling uncertainty of this ratio, pressures greater than ~750 bars (Kennedy, 1950; Fig. 15A).
we conclude that 100 g input fluid of intermediate density If the fluid is saline, this normal solubility dependence ex-
separated into 90 3 g vapor and 10 3 g brine during as- tends to lower pressures, as indicated by the gray isopleths in
cent. To achieve a plausible fit, the starting salinity of the in- Figure 15A (Fournier, 1983). Quartz solubility in low-salinity
termediate-density fluids had to be 7 2 wt percent NaCl fluids at low pressure shows a region of inverse temperature
equiv, corresponding to 8 to 13 wt percent total dissolved dependence, i.e., quartz solubility increases upon cooling
salts. Total salinity of the vapor ascending through the central from ~450 to ~350C. We infer that most of Q1 in stockwork
Au-Cu orebody is computed to be about half the salinity of veins in and below the orebody probably precipitated upon
the single-phase input fluid, whereas in the periphery the cooling and decompression from >500C and 800 bars to
vapor is closer in salinity to the input fluid. The higher pre- lower pressures and temperatures, as the fluids were ejected
dicted vapor salinity in the Cu-rich periphery and the lower from the subjacent lithostatic fluid-pressure domain into an
salinity of the associated brines (determined by microther- environment of hot hydrostatic conditions (Gruen et al.,
mometry, Fig. 9) are consistent with the phase proportion sta- 2010). Based on equilibrium quartz solubility variation, this
tistics (Fig. 7), indicating a narrower two-phase region at process will deposit 0.1 to 0.2 g of vein quartz per 100 g of hy-
slightly higher pressure and/or lower temperature compared drothermal fluid. Fluid packets experiencing decompression
to the Au-Cu center of the system. Element/Na ratios in the from 100 to 300 bars prior to cooling below ~450C, as
predicted vapor agree with the vapor inclusion analyses recorded by most of the fluid inclusion assemblages (Fig. 9),
within their overall variability, with regard to all well-defined will pass through the region of retrograde solubility and
elements including the main ore metals Cu, Mo, and Au. therefore locally redissolve quartz. This explains the creation
Prior to the onset of Cu-Fe sulfide precipitation, the vapor of secondary porosity in Q1 veins, prior to recementation with
contains Cu as its most abundant metal component, in agree- Q2 quartz (Fig. 5). In the two-phase region of brine + vapor
ment with the most copper rich assemblages in the broad coexistence, net quartz dissolution will result only if the low-
range of analyses. The high mass fraction of vapor and its high salinity vapor predominates over the high-salinity brine; the
Cu content imply that 95 percent of the total copper was de- brine by itself has no retrograde solubility even at low pres-
posited by the vapor phase. In the periphery, the observed Cu sure (e.g., gray isopleth labeled 20% NaCl, Fig. 15A). We
content of the inclusions is smaller than the prediction, prob- have not tried to quantify the combined effects of adiabatic
ably reflecting the smaller data set and the fact that chalcopy- expansion, cooling, and quartz solubility in a two-phase saline
rite is already precipitating at greater depth, possibly prior to fluid mixture (cf. Akinfiev and Diamond, 2009), but the quartz
phase separation (Figs. 10, 11). Detectable gold in intermedi- dissolution textures are an independent qualitative confirma-
ate-density and vapor fluid inclusions from the deep periph- tion that ore deposition in the Bingham Canyon system was
ery and from the Cu-rich but Au-poor peripheral ore zone in- dominated by magmatic vapor, rather than saline liquid.
dicates that gold is transported through the peripheral upflow
region, without precipitating within the current mine volume. Fluid pressure control on Au/Cu ore-grade zonation
The physical process of phase separation depends on The main difference between the Au-Curich center and
whether the input fluid breaches the two-phase surface on the relatively Cu-rich periphery of the system is in the slightly
0361-0128/98/000/000-00 $6.00 113

vapor density is the stabilization of metal ions and neutral

A Quartz solubility 0.2
r
complexes by hydration, i.e., by the formation of a shell of
ba
00
30 water molecules surrounding the metal-carrying species in
millimol SiO 2 / kg H2O
10
20
Wt % SiO 2
the vapor. Although such experiments have not yet been com-
750 pleted for complex vapors in the Au-Cu-Fe-S-(O-H) system
15
salinity
(cf. Zezin et al., 2008), the strong decompression and expan-
NaCl % sion of magmatic vapor ascending through the center of the
20
10 fluid plume at Bingham Canyon is expected to lead to desta-
0.1
bilization of transported metals, resulting in wholesale pre-
50 cipitation of most of the advected Cu and Au.
5 0
ba These experiments indicate that the most Au enriched and
r
10 the highest grade center of the orebody was formed in the re-
psat 2
gion of greatest fluid decompression. Indeed, the bulk Au/Cu
HO
ratio of this region approximately matches that of the input
300 bar2 fluid composition, consistent with complete deposition of all
advected copper and gold. Vapor expansion and separation of
0 0 minor brine will be associated with adiabatic cooling of the
fluid mixture. Indeed, a steep decrease of Cu content of as-
0.5
sociated brine inclusions is correlated with decreasing ho-
B Copper content
Cu / Na (by wt)
mogenization temperature of brine inclusion assemblages

Secondary brine inclusions
in Q1 quartz
(Fig. 15B; modified from Landwing et al., 2005). Figure 10
0.1
Brine inclusions
(sample D211-19) shows that the vapor inclusions also record
in Q2 quartz a decrease in Cu content over several orders of magnitude,
starting from an even higher Cu/Na ratio compared with the
brines (Fig. 11). Note that the temperature range of Cu-Fe
sulfides precipitatation from ~430 to ~350C (Fig. 15) is too
0.01 low to initially incorporate all the gold in bornite solid solu-
tion (Simon et al., 2000; Redmond, 2002; cf. Kesler et al.,
2002). This implies that native gold coprecipitated with cop-
per iron sulfides in the Au-Curich center of the orebody.
Fluids ascending along the periphery of the fluid plume re-
0.001
mained denser, which favors the retention of Au in solution
200 300 400 500
while copper is precipitating selectively. Even though solubil-
Temperature C ity experiments are lacking to demonstrate this proposed
FIG. 15. Experimental quartz solubility as a function of pressure, temper- metal separation, thermodynamic modeling of intermediate-
ature, and fluid salinity (A, based on data from Kennedy, 1950, and Fournier, density fluids and aqueous liquids (Heinrich, 2005) predict
1983) and the observed evolution of copper content as a function of homog- rapidly decreasing chalcopyrite solubility in any cooling S-rich
enization temperature in brine inclusion assemblages from sample D211-19 fluid, while gold solubility remains high due to the formation
(B, redrawn in terms of element ratios from data in Landtwing et al., 2005,
and digital supplement). The arrows indicate how redissolution of the previ- of stable Au-HS complexes (Gammons and Williams-Jones,
ously deposited main generation of vein quartz (Q1; cf. Fig. 5) and Cu pre- 1997; Stefnsson and Seward, 2004). High sulfur concentra-
cipitation have occurred together in the same cooling interval, from a pre- tions in the magmatic vapor at Bingham Canyon were re-
dominant low-salinity fluid of vaporlike density. cently confirmed by fluid inclusion analysis (Seo et al., 2009).
Thus, we propose that the relatively gold poor but Cu rich pe-
riphery of the Bingham Canyon orebody formed by selective
different fluid evolution paths, which led to more extreme precipitation of chalcopyrite, while most of the advected gold
phase separation in the center compared to the periphery of was transported through and out of the porphyry orebody.
the orebody. Volume expansion, phase separation, adiabatic Upon further cooling, the dense vapor can contract to an
cooling, and changing mineral solubilities in coexisting brine aqueous liquid and may eventually contribute to the forma-
and vapor are all linked with each other but cannot be fully tion of distal epithermal gold deposits such as Melco or Bar-
quantified for lack of thermodynamic data. Nevertheless, the neys Canyon (Sillitoe and Bonham, 1990; Cunningham et al.,
combined observations and limited new experimetal data in- 2004). These deposits are small, but our interpretation im-
dicate that small differences in the density evolution of the plies that a much greater quantity of gold may have been lost
cooling fluids could explain the zoned Au/Cu ratio in the from the system. This could amount to ~1,400 metric tons of
Bingham Canyon orebody and the particularly high ore lost gold if we assume that the Au/Cu ratio of the central
grades in the center of the system. Au-Cu orebody (0.8 104) represents the bulk metal ratio of
Experimental measurements of the solubility of metallic the input fluid and consider the known reserves of additional
minerals in hydrous vapor have shown a strong positive cor- 19 Mt copper in Au-poor peripheral ore zones (Gruen et al.,
relation with the density of the vapor (Williams-Jones et al., 2010).
2002; Williams-Jones and Heinrich, 2005). At the atomic scale, Later fluids forming the quartz-molybdenite veins are com-
the reason for increasing metal solubility with increasing positionally similar to the quartz stockwork veins associated
0361-0128/98/000/000-00 $6.00 114

with Cu-Au mineralization (Figs. 10, 11), although some in-

clusion assemblages have slightly higher Mo/Na ratios. They
seem to have experienced a similar pressure-temperature-
fumarole
phase evolution, without intermitttent cooling to the lower
temperatures characterizing the quartz-pyrite veins with
sericitic alteration. The latter contain aqueous inclusions with
high Cu and Au contents, comparable to the transitional high- increasing
sulfidation epithermal fluids observed at La Famatina (Ar- interaction
gentina; Pudack et al., 2009). These fluids, too, are potential and mixing
source fluids for distal epithermal deposits. with
VOLC
Fluid flow and hydrology
surface water...
The variation in fluid properties derived from inclusions in vapor
the quartz stockwork veins allow a qualitative interpretation expansion
of the flow pattern during Cu-Au mineralization at Bingham vapor vapor
Canyon, consistent with the rock mechanical concept devel- contraction contraction
oped by Gruen et al. (2010). Figure 16 captures a schematic
snapshot of a zoned magmatic fluid plume, indicating the brine
varying density of vapor, intermediate-density fluids and liq- or deeply
SED
condensation
uids by variably dark shading. This cartoon is a gross simplifi- convecting
cation of a dynamic and locally very heterogeneous fluid-flow basin
system. However, the observed large-scale mineral and inclu- fluids
sion zoning indeed requires a period of relatively stationary
distribution of fluid pressure and temperature over several
cycles of stockwork veining and Cu-Au mineralization.
A steep vertical gradient in fluid pressure and any lateral EM
pressure difference between the center and the periphery of C
0
40
the fluid plume cannot be maintained in a stagnant hydrosta-
tic environment. However, they can persist within a dynamic broad source of overpressured
fluid plumeeven under the extreme assumptions that rock single-phase magmatic fluid QMP
permeability was similar across a broad upflow zone and that
us)
fluid pressure in the ore-depositing region was determined by T(solid
the weight of the locally overlying fluid column. Such con-
vective flow is predicted by hydrodynamic modeling (Dries- FIG. 16. Cartoon of the magmatic fluid plume that may have generated
ner and Geiger, 2007), indicating that the center of a vapor- the zoned ore shell of the Bingham Canyon porphyry Cu-Au deposit, em-
phasizing the observed variations in fluid density by variable degrees of shad-
dominated two-phase fluid plume can maintain a relatively ing from low-density steam (white) to dense brine (black). The region of
low pressure approaching vapor-static conditions. Vapor static vapor density variation and local brine condensation is dominated by mag-
means that pressure is determined dominantly by the inte- matic fluids, whereas the more evenly dark-gray surrounding may be domi-
grated weight of upward-expanding low-density vapor and nated by surface-derived aqueous liquids. Interaction between magmatic and
meteoric fluid is most likely in the near-surface fumarole environment where
possibly overlying meteoric water. At the same time, the pe- vapor will condense into ground water or at the periphery where contracted
riphery of the magmatic fluid plume can be closer to hydro- magmatic vapor becomes completely miscible with external aqueous liquids.
static condition, controlled by the weight of a more slowly Abbreviations: EM = equigranular monzonite, QMP = quartz monzonite
flowing fluid mixture with overall higher density. The higher porphyry, SED = sedimentary rocks, VOLC = volcanic rocks.
fluid density in the periphery of the plume is stabilized by two
effects that mutually enhance each other: magmatic fluids
cooling under elevated pressure maintains its density or even ore mineral saturation front within the overlying low-pressure
contracts to liquid (Heinrich et al., 2004), and elevated pres- domain, which probably remained relatively stationary for an
sure is in turn maintained by the increasing density of the extended period while fluids were continually advecting heat,
magmatic fluid, as well as liquid meteoric water convecting sulfur, and metals through the vein network. The basal front
above the periphery of the fluid plume. of Cu-Au saturation probably approximates an isotherm sur-
In reality, fluid pressure will additionally be affected by face near 430C, based on the temperature of initial copper
variable rock permeability. Fracture permeability varies due sulfide deposition indicated by Figure 15B (Landtwing et al.,
to hydraulic fracturing at the transition from near-lithostatic 2005). The thermal balance between fluid advection and heat
pressure conditions at depth, to pressure conditions closer to loss required to maintain such a stationary saturation front is
the Earths surface that are determined by fluid density. probably a key factor for generating the exceptionally high Cu
Lithostatic fluid pressure prevails in the underlying source and Au grades in the center of the Bingham Canyon orebody.
magma (below the solidus isotherm indicated schematically Better understanding this dynamic stability in an inherently
in Fig. 16) and probably extends upward into the broad re- fluctuating high-energy environment will require quantitative
gion characterized by intermediate-density inclusions. The evaluation of interacting physical and chemical processes
sharp base of the central Au-Cu orebody is interpreted as an (Steinberger et al., 2009; Weis et al., 2009).
0361-0128/98/000/000-00 $6.00 115

The now eroded surface environment at the time of Cu-Au cooling magmatic vapor. The nonprecipitating gold would as-
mineralization is open to speculation. The vapor plume may cend in the contracting vapor and be potentially available to
have reached the paleosurface as a giant fumarole or the form epithermal deposits at much shallower depth.
vapor may eventually have mixed with near-surface water,
e.g., in a laterally draining crater lake. Once copper-depleted Acknowledgments
but still gold-rich magmatic vapor ascending along the pe- We would like to thank the geology staff at Kennecott Utah
riphery of the fluid plume had contracted to an aqueous liq- Copper (KUC) and their exploration colleagues in Rio Tinto,
uid below ~350C, it would have become completely miscible who hosted us during many visits at the Bingham Canyon
with meteoric water. The mixed aqueous solution could have mine. Special thanks go to Gerry Austin, Geoff Ballantyne,
flowed laterally and downward depending on surface topog- Mike Pennick, John Porter, Martyn Robotham, Kim Schroeder,
raphy to contribute to the formation of sediment-hosted gold Tracy Smith, and the late Ricardo Presnell for their help,
deposits. A downward-mineralizing flow path has been pro- stimulating discussions, and useful comments on these stud-
posed by Cunningham et al. (2004) for the Melco deposit, lo- ies and papers. We are most grateful to Marco Einaudi for
cated about 5 km north of Bingham Canyon, because this sed- having laid the groundwork to this joint scientific effort and to
iment-hosted epithermal deposit is conspicuously quartz poor him as well as Thomas Driesner, John Dilles, Werner Halter,
and has no discernible fluid feeder zone in its footwall. Dick Henley, Ken Krahulec, Jeff Keith, Philipp Weis, An-
thony Williams-Jones, and many other colleagues and fellow
Conclusions students for their comments and ideas. This research was
Geologic observations and fluid inclusion properties at supported by ETH project 0-20663-99 and continuing sup-
Bingham Canyon extend the classic vapor plume model for port from ETH Zurich and the Swiss National Science Foun-
magmatic-hydrothermal ore systems (Henley and McNabb, dation (project 200020-116639).
1978) with specific conclusions regarding the hydrothermal
geochemistry of porphyry Cu-Au ore deposition (cf. Williams- REFERENCES
Jones and Heinrich, 2005). Akinfiev, N.N., and Diamond, L.W., 2009, A simple predictive model of
quartz solubility in water-salt-CO2 systems at temperatures up to 1000 C
The chemical mineralization efficiency and ore grade of and pressures up to 1000 MPa: Geochimica et Cosmochimica Acta, v. 73,
porphyry Cu Au deposits is probably dominated by the sul- p. 15971608.
fur chemistry of low-salinity magmatic fluids and the density Anderson, A.J., Clark, A.H., Ma, X.P., Palmer, G.R., Macarthur, J.D., and
evolution path of the fluids during cooling. Fluid inclusion Roedder, E., 1989, Proton-induced X-ray and gamma-ray emission analysis
analyses presented in this paper and by Rusk et al. (2004, of unopened fluid inclusions: ECONOMIC GEOLOGY, v. 84, p. 924939.
Archibald, S.M., Migdisov, A.A., and Williams-Jones, A.E., 2002, An experi-
2008; Butte), Klemm et al. (2007; El Teniente) and Zajacz and mental study of the stability of copper chloride complexes in water vapor at
Halter (2009; phenocryst inclusions in recent andesite) are in elevated temperatures and pressures: Geochimica et Cosmochimica Acta,
line with recently published experimental data on sulfide sol- v. 66, p. 16111619.
ubility and fluid partitioning (Nagaseki and Hayashi, 2008; Audtat, A., and Gnther, D., 1999, Mobility and H2O loss from fluid inclu-
Pokrovski et al., 2008), and with the discovery of high sulfur sions in natural quartz crystals: Contributions to Mineralogy and Petrology,
v. 137, p. 114.
concentrations in metal-rich vapor inclusions (Seo et al., Audtat, A., Gnther, D., and Heinrich, C.A., 1998, Formation of a mag-
2009). These data consistently indicate that the high Cu and matic-hydrothermal ore deposit: Insights with LA-ICP-MS analysis of fluid
Au concentrations found in many low-salinity magmatic fluids inclusions: Science, v. 279, p. 20912094.
are controlled by sulfur as the dominant complexing ligand, Audtat, A., Pettke, T., Heinrich, C.A., and Bodnar, R.J., 2008, The compo-
sition of magmatic-hydrothermal fluids in barren and mineralized intru-
rather than by chloride complexation as previously assumed. sions: ECONOMIC GEOLOGY, v. 103, p. 877908.
The bulk Au/Cu ratio of porphyry-style ore deposits may be Baker, T., Van Achterberg, E., Ryan, C.G., and Lang, J.R., 2004, Composi-
controlled at least partly by the vapor evolution process that tion and evolution of ore fluids in a magmatic-hydrothermal skarn deposit:
explains the Au and Cu zoning within the porphyry ore shell Geology, v. 32, p. 117120.
at Bingham Canyon. The Au/Cu ratio of porphyry deposits Bodnar, R.J., 1995, Fluid-inclusion evidence for a magmatic source of metals
in porphyry copper deposits: Mineralogical Association of Canada Short
empirically correlates with the depth and fluid pressure at- Course Series 23, p. 139152.
tending ore deposition (Cox and Singer, 1988; Sillitoe, 1997), Bodnar, J.R., and Vityk, M.O., 1994, Interpretation of microthermometric
as discussed by Murakami et al. (2009). Au-rich porphyry de- data for H2O-NaCl fluid inclusions, in De Vivo, B., and Frezzotti, M.L.,
posits in which Au and Cu coprecipitated at high temperature eds., Fluid iInclusions in minerals: Methods and applications: Blacksburg,
VA, Virginia Technical Institute, p. 117130.
are typically formed at low fluid pressures in the subvolcanic Candela, P.A., and Holland, H.D., 1986, A mass-transfer model for copper
domain, for example, Grasberg (MacDonald and Arnold, and molybdenum in magmatic hydrothermal systemsthe origin of por-
1994), Bajo de la Alumbrera (Ulrich et al., 2002), and the porphyry-type ore deposits: ECONOMIC GEOLOGY, v. 81, p. 119.
phyry gold deposits of the Maricunga belt (Vila and Sillitoe, Candela, P.A., and Piccoli, P.M., 1995, Model ore-metal partitioning from
1991). By contrast, more deeply seated Cu Mo deposits are melts into vapor and vapor/brine mixtures: Mineralogical Association of
Canada Short Course Series 23, p. 101127.
typically gold poor, for example El Teniente (formed from a Cathles, L.M., 1977, An analysis of the cooling of intrusives by ground-water
low-salinity vapor intersecting the two-phase region near the convection which includes boiling: ECONOMIC GEOLOGY, v. 72, p. 804826.
critical curve or contracting directly to aqueous liquid; Cline, J.S., and Bodnar, R.J., 1991, Can economic porphyry copper mineral-
Klemm et al., 2007), or Butte (where ore deposition occurred ization be generated by a typical calc-alkaline melt: Journal of Geophysical
at high pressure from a single-phase fluid; Rusk et al., 2004, Research-Solid Earth and Planets, v. 96, p. 81138126.
Cox, D.P., and Singer, D.A., 1988, Distribution of gold in porphyry copper
2008). Our model implies that, even if the input fluids into deposits: U.S. Geological Survey Open-File Report 8846, 23 p.
some of these Au-poor deposits contained some gold, copper Cunningham, C.G., Austin, G.W., Naeser, C.W., Rye, R.O., Ballantyne,
was precipitated selectively due to the higher density of the G.H., Stamm, R.G., and Barker, C.E., 2004, Formation of a paleothermal
0361-0128/98/000/000-00 $6.00 116

anomaly and disseminated gold deposits passociated with the Bingham Heinrich, C.A., Ryan, C.G., Mernagh, T.P., and Eadington, P.J., 1992, Segre-
Canyon Porphyry Cu-Au-Mo system, Utah: ECONOMIC GEOLOGY, v. 99, p. gation of ore metals between magmatic brine and vapora fluid inclusion
789806. study using PIXE microanalysis: ECONOMIC GEOLOGY, v. 87, p. 15661583.
Damman, A.H., Kars, S.M., Touret, J.L.R., Rieffe, E.C., Kramer, J., Vis, Heinrich, C.A., Gnther, D., Audtat, A., Ulrich, T., and Frischknecht, R.,
R.D., and Pintea, I., 1996, PIXE and SEM analyses of fluid inclusions in 1999, Metal fractionation between magmatic brine and vapor, determined
quartz crystals from the K-alteration zone of the Rosia Poieni porphyry-Cu by microanalysis of fluid inclusions: Geology, v. 27, p. 755758.
deposit, Apuseni mountains, Rumania: European Journal of Mineralogy, v. Heinrich, C.A., Pettke, T., Halter, W.E., Aigner-Torres, M., Audtat, A., Gun-
8, p. 10811096. ther, D., Hattendorf, B., Bleiner, D., Guillong, M., and Horn, I., 2003,
Deino, A., and Keith, J.D., 1997, Ages of volcanic and intrusive rocks in the Quantitative multi-element analysis of minerals, fluid and melt inclusions
Bingham mining district, Utah: Society of Economic Geologists Guidebook by laser ablation-inductively coupled plasma-mass spectrometry: Geochim-
Series, v. 29, p. 91100. ica et Cosmochimica Acta, v. 67, p. 34733497.
Denis, M., Pichavant, M., Poty, B., and Weisbrod, A., 1980, The Sierrita-Es- Heinrich, C.A., Driesner, T., Stefnsson, A., and Seward, T.M., 2004, Mag-
peranza porphyry copper, Arizona, USA: Bulletin de Mineralogie, v. 103, p. matic vapor contraction and the transport of gold from the porphyry envi-
613622. ronment to epithermal ore deposits: Geology, v. 32, p. 761764.
Diamond, L.W., 1992, Stability of CO2 clathrate hydrate + CO2 liquid + CO2 Henley, R.W., and McNabb, A., 1978, Magmatic vapor plumes and ground-
vapour + aqueous KCl-NaCl solutionsexperimental-determination and water interaction in porphyry copper emplacement: ECONOMIC GEOLOGY,
application to salinity estimates of fluid inclusions: Geochimica et Cos- v. 73, p. 120.
mochimica Acta, v. 56, p. 273280. Inan, E.E., 2002, Occurrence of nukundamite at the Bingham porphyry Cu-
Dilles, H.J., 1987, Petrology of the Yerington batholith, Nevada: Evidence for Au-Mo district, Utah: Unpublished Ph.D. thesis, Stanford, CA, Stanford
evolution of porphyry copper ore fluids: ECONOMIC GEOLOGY, v. 82, p. University, 221 p.
17501789. Irwin, J.J., and Roedder, E., 1995, Diverse origins of fluid in magmatic in-
Driesner, T., and Geiger, S., 2007, Numerical simulation of multiphase fluid clusions at Bingham (Utah, USA), Butte (Montana, USA), St-Austell
flow in hydrothermal systems: Reviews in Mineralogy and Geochemistry, v. (Cornwall, UK), and Ascension-Island (Mid-Atlantic, UK), indicated by
65, p. 187212. laser microprobe analysis of Cl, K, Br, I, Ba Plus Te, U, Ar, Kr, and Xe:
Driesner, T., and Heinrich, C.A., 2007, The system H2O-NaCl. Part I: Cor- Geochimica et Cosmochimica Acta, v. 59, p. 295312.
relation formulae for phase relations in temperature-pressure-composition Kendrick, M.A., Burgess, R., Pattrick, R.A.D., and Turner, G., 2001a, Fluid
space from 0 to 1000C, 0 to 5000 bars, and 0 to 1 X-NaCl: Geochimica et inclusion noble gas and halogen evidence on the origin of Cu-porphyry
Cosmochimica Acta, v. 71, p. 48804901. mineralising fluids: Geochimica et Cosmochimica Acta, v. 65, p.
Eastoe, C.J., 1978, A fluid inclusion study of the Panguna porphyry copper 26512668.
deposit, Bougainville, Papua New Guinea: ECONOMIC GEOLOGY, v. 73, p. 2001b, Halogen and Ar-Ar age determinations of inclusions within
721748. quartz veins from porphyry copper deposits using complementary noble
Etminan, H., 1977, Le porphyre cuprifere de Sar Cheshmeh (Iran): Role des gas extraction techniques: Chemical Geology, v. 177, p. 351370.
phases fluides dans ales mecanismes d`alteration et de mineralisation: Un- Kennedy, G.C., 1950, A portion of the system silica-water: ECONOMIC GEOL-
published Ph.D. thesis, Nancy, France, University of Nancy, 249 p. OGY, v. 45, p. 629653.
Fournier, R.O., 1983, A method of calculating quartz solubilities in aqueous Kesler, S.E., Chryssoulis, S.L., and Simon, G., 2002, Gold in porphyry copper
sodium-chloride solutions: Geochimica et Cosmochimica Acta, v. 47, p. deposits: Its abundance and fate: Ore Geology Reviews, v. 21, p. 103124.
579586. Klemm, L.M., Pettke, T., Heinrich, C.A., and Campos, E., 2007, Hydrother-
Furrer, C., 2006, Fluid evolution and metal zonation at the Bingham por- mal evolution of the El Teniente deposit, Chile: Porphyry Cu-Mo ore de-
phyry Cu-Au-Mo deposit, Utah: Constraints from microthermometry and position from low-salinity magmatic fluids: ECONOMIC GEOLOGY, v. 102, p.
LA-ICPMS analysis of fluid inclusions: Unpublished M.Sc. thesis, Zurich, 10211045.
Switzerland, ETH Zurich, 96 p. 2008, Fluid and source magma evolution of the Questa porphyry Mo de-
Gammons, C.H., and Williams-Jones, A.E., 1997, Chemical mobility of gold in posit, New Mexico, USA: Mineralium Deposita, v. 43, p. 533552.
the porphyry-epithermal environment: ECONOMIC GEOLOGY, v. 92, p. 4559. Landtwing, M.R., 2004, Fluid evolution and ore mineral precipitation at the
Gehrig, M., 1980, Phasengleichgewichte und PVT-Daten ternrer Mischun- Bingham porphyry Cu-Au-Mo deposit, Utah, deduced from cathodolumi-
gen aus Wasser, Kohledioxid und Natriumchlorid bis 3 kbar und 550C: nescence imaging and LA-ICPMS microanalysis of fluid inclusions: Un-
Unpublished Ph.D. thesis, Germany, University of Karlsruhe, 109 p. published Ph.D. thesis, Zurich, Switzerland, ETH Zurich, 260 p.
Giggenbach, W.F., 1992, Magma degassing and mineral deposition in hy- Landtwing, M.R., and Pettke, T., 2005, Relationships between SEM-
drothermal systems along convergent plate boundaries: ECONOMIC GEOL- cathodoluminescence response and trace-element composition of hy-
OGY, v. 87, p. 19271944. drothermal vein quartz: American Mineralogist, v. 90, p. 122131.
Goldstein, R.H., and Reynolds, T.J., 1994, Systematics of fluid inclusions in Landtwing, M.R., Pettke, T., Halter, W.E., Heinrich, C.A., Redmond, P.B.,
diagenetic minerals: Society for Sedimentary Geology, 199 p. Einaudi, M.T., and Kunze, K., 2005, Copper deposition during quartz dis-
Gruen, G., 2007, Distribution and orientation of veins in the Bingham solution by cooling magmatic-hydrothermal fluids: The Bingham porphyry:
Canyon porphyry Cu-Mo-Au deposit, Utah: Unpublished M.Sc. thesis, Earth and Planetary Science Letters, v. 235, p. 229243.
Zurich, Switzerland, ETH Zurich, 123 p. Li, Y., Audtat, A., Lerchbaumer, L., and Xiong, X.L., 2009, Rapid Na, Cu ex-
Gruen, G., Heinrich, C.A., and Schroeder, K., 2010, The Bingham Canyon change between synthetic fluid inclusions and external aqueous solutions:
porphyry Cu-Mo-Au deposit. II. Vein geometry and ore shell formation by Evidence from LA-ICP-MS analysis: Geofluids, v. 9, p. 321329.
pressure-driven rock extension: ECONOMIC GEOLOGY, v. 105, p. 6990. Liebscher, A., 2007, Experimental studies in model fluid systems: Reviews in
Gnther, D., Frischknecht, R., Heinrich, C.A., and Kahlert, H.J., 1997, Ca- Mineralogy and Geochemistry, v. 65, p. 1547.
pabilities of an argon fluoride 193 nm excimer laser for laser ablation in- Liebscher, A., and Heinrich, C.A., 2007, Fluid-fluid interactions in the earths
ductively coupled plasma mass spectrometry microanalysis of geological lithosphere: Fluid-Fluid Interactions, v. 65, p. 113.
materials: Journal of Analytical Atomic Spectrometry, v. 12, p. 939944. Lindgren, W., 1905, The copper deposits of the Clifton-Morenci district, Ari-
Gnther, D., Audtat, A., Frischknecht, R., and Heinrich, C.A., 1998, Quan- zona: U.S. Geological Survey Professional Paper 43, 375 p.
titative analysis of major, minor and trace elements in fluid inclusions using Lowenstern, J.B., Mahood, C.A., Rivers, M.L., and Sutton, S.R., 1991, Evi-
laser ablation inductively coupled plasma mass spectrometry: Journal of dence for extreme partitioning of copper into a magmatic vapor phase: Sci-
Analytical Atomic Spectrometry, v. 13, p. 263270. ence, v. 252, p. 14051409.
Guillong, M., Latkoczy, C., Seo, J.H., Gunther, D., and Heinrich, C.A., 2008, MacDonald, G.D., and Arnold, L.C., 1994, Geological and geochemical zon-
Determination of sulfur in fluid inclusions by laser ablation ICP-MS: Jour- ing of the Grasberg Igneous Complex, Irian-Jaya, Indonesia: Journal of
nal of Analytical Atomic Spectrometry, v. 23, p. 15811589. Geochemical Exploration, v. 50, p. 143178.
Hedenquist, J.W., and Lowenstern, J.B., 1994, The role of magmas in the for- Murakami, H., Seo, J.H., and Heinrich, C.A., 2009, The relation between
mation of hydrothermal ore-deposits: Nature, v. 370, p. 519527. Cu/Au ratio and formation depth of porphyry-style Cu-Au Mo deposits:
Heinrich, C.A., 2005, The physical and chemical evolution of low-salinity Mineralium Deposita, DOI10.1007/s0012600902551.
magmatic fluids at the porphyry to epithermal transition: A thermodynamic Nagaseki, H., and Hayashi, K.I., 2008, Experimental study of the behavior of
study: Mineralium Deposita, v. 39, p. 864889. copper and zinc in a boiling hydrothermal system: Geology, v. 36, p. 2730.
0361-0128/98/000/000-00 $6.00 117

Nash, J.T., 1976, Fluid-inclusion petrology: Data from porphyry copper de- Sillitoe, R.H., and Bonham, H.F., 1990, Sediment-hosted gold depositsdis-
posits and applications to exploration: U.S. Geological Survey Professional tal products of magmatic-hydrothermal systems: Geology, v. 18, p. 157161.
Paper, 907 D, p. 116. Simon, A.C., Pettke, T., Candela, P.A., Piccoli, P.M., and Heinrich, C.A., 2007,
Pettke, T., Heinrich, C.A., Ciocan, A.C., and Gnther, D., 2000, Quadrupole The partitioning behavior of As and Au in S-free and S-bearing magmatic
mass spectrometry and optical emission spectroscopy: Detection capabili- assemblages: Geochimica et Cosmochimica Acta, v. 71, p. 17641782.
ties and representative sampling of short transient signals from laser-abla- Simon, G., Kesler, S.E., Essene, E.J., and Chryssoulis, S.L., 2000, Gold in
tion: Journal of Analytical Atomic Spectrometry, v. 15, p. 11491155. porphyry copper deposits: Experimental determination of the distribution
Pokrovski, G.S., Roux, J., and Harrichoury, J.C., 2005, Fluid density control of gold in the Cu-Fe-S system at 400 to 700C: ECONOMIC GEOLOGY, v. 95,
on vapor-liquid partitioning of metals in hydrothermal systems: Geology, v. p. 259270.
33, p. 657660. Stefnsson, A., and Seward, T.M., 2004, Gold(I) complexing in aqueous sul-
Pokrovski, G.S., Borisova, A.Y., and Harrichoury, J.C., 2008, The effect of sul- phide solutions to 500C at 500 bar: Geochimica et Cosmochimica Acta, v.
fur on vapor-liquid fractionation of metals in hydrothermal systems: Earth 68, p. 41214143.
and Planetary Science Letters, v. 266, p. 345362. Steinberger, I., Driesner, T., Weis, P., and Heinrich, C.A., 2009, Towards a
Pudack, C., Halter, W.E., Heinrich, C.A., and Pettke, T., 2009, Evolution of numerical reconstruction of the Bingham Canyon magmatic-hydrothermal
magmatic vapor to gold-rich epithermal liquid: The porphyry to epithermal ore system [abs.]: Geochimica et Cosmochimica Acta, v. 73, p. A1269.
transition at Nevados de Famatina, northwest Argentina: ECONOMIC GE- Taylor, H.P.J., 1974, The application of oxygen and hydrogen isotope studies
OLOGY, v. 104, p. 449477. to problems of hydrothermal alteration and ore deposition: ECONOMIC GE-
Redmond, P.B., 2002, Magmatic-hydrothermal fluids and copper-gold ore OLOGY, v. 69, p. 843883.
formation at Bingham Canyon, Utah: Unpublished Ph.D. thesis, Stanford, Ulrich, T., Gnther, D., and Heinrich, C.A., 1999, Gold concentrations of
CA, Stanford University, 228 p. magmatic brines and the metal budget of porphyry copper deposits: Na-
Redmond, P.B., and Einaudi, M.T., 2010, The Bingham Canyon porphyry ture, v. 399, p. 676679.
Cu-Mo-Au deposit. I. Sequence of intrusions, vein formation, and sulfide 2002, The evolution of a porphyry Cu-Au deposit, based on LA-ICP-MS
deposition: ECONOMIC GEOLOGY, v. 105, p. 4368. analysis of fluid inclusions: Bajo de la Alumbrera, Argentina (vol 96, pg
Redmond, P.B., Landtwing, M.R., and Einaudi, M.T., 2001, Cycles of por- 1743, 2001): ECONOMIC GEOLOGY, v. 97, p. 18881920.
phyry dike emplacement, veining, alteration and mineralisation in the Vanko, D.A., Bonnin-Mosbah, M., Philippot, P., Roedder, E., and Sutton,
Bingham porphyry Cu-Au-Mo deposit, Utah, in Piestrzyski, A., et al., S.R., 2001, Fluid inclusions in quartz from oceanic hydrothermal speci-
Mineral deposits at the beginning of the 21st century: Netherlands, mens and the Bingham, Utah porphyry-Cu deposit: A study with PIXE and
A.A.Balkema, p. 473476. SXRF: Chemical Geology, v. 173, p. 227238.
Redmond, P.B., Einaudi, M.T., Inan, E.E., Landtwing, M.R., and Heinrich, Vila, T., and Sillitoe, R.H., 1991, Gold-rich porphyry systems in the Mari-
C.A., 2004, Copper deposition by fluid cooling in intrusion-centered sys- cunga belt, northern Chile: ECONOMIC GEOLOGY, v. 86, p. 12381260.
tems: New insights from the Bingham porphyry ore deposit, Utah: Geology, Waite, K.A., Keith, J.D., Christiansen, E.H., Whitney, J.A., Hattori, K.H.,
v. 32, p. 217220. Tingey, D.G., and Hook, C.J., 1997, Petrogenesis of the volcanic and intru-
Roedder, E., 1971, Fluid inclusion studies on the porphyry-type ore deposits sive rocks associated with the Bingham Canyon porphyry Cu-Au-Mo de-
at Bingham, Utah, Butte, Montana, and Climax, Colorado: ECONOMIC GE- posit: Society of Economic Geologists Guidebook Series, v. 29, p. 6990.
OLOGY, v. 66, p. 98120. Weis, P., Driesner, T., Heinrich, C.A., Coumou, D., and Geiger, S., 2009,
1992, Fluid inclusion evidence for immiscibility in magmatic differenti- Flow of brine and vapour in subaerial and submarine magmatic hydrother-
ation: Geochimica et Cosmochimica Acta, v. 56, p. 520. mal systems: SGA Conference, Townsville, August 2009, Proceedings, p.
Rusk, B.G., and Reed, M.H., 2002, Scanning electron microscope-cathodo- 873875.
luminescence analysis of quartz reveals complex growth histories in veins Williams, T.J., Candela, P.A., and Piccoli, P.M., 1995, The partitioning of cop-
from the Butte porphyry copper deposit, Montana: Geology, v. 30, p. per between silicate melts and two-phase aqueous fluids: An experimental
727730. investigation at 1 kbar, 800C and 0.5 kbar, 850C: Contributions to Min-
Rusk, B.G., Reed, M.H., Dilles, J.H., Klemm, L.M., and Heinrich, C.A., eralogy and Petrology, v. 121, p. 388399.
2004, Compositions of magmatic hydrothermal fluids determined by LA- Williams-Jones, A.E., and Heinrich, C.A., 2005, Vapor transport of metals
ICP-MS of fluid inclusions from the porphyry copper-molybdenum deposit and the formation of magmatic-hydrothermal ore deposits: ECONOMIC GE-
at Butte, MT: Chemical Geology, v. 210, p. 173199. OLOGY, v. 100, p. 12871312.
Rusk, B.G., Reed, M.H., and Dilles, J.H., 2008, Fluid inclusion evidence for Williams-Jones, A.E., Migdisov, A.A., Archibald, S.M., and Xiao, Z.F., 2002,
magmatic-hydrothermal fluid evolution in the porphyry copper-molybde- Vapor-transport of ore metals: Gochemical Society Special Publication 7, p.
num deposit at Butte, Montana: ECONOMIC GEOLOGY, v. 103, p. 307334. 279305.
Sawkins, F.J., and Scherkenbach, D.A., 1981, High copper content of fluid Zajacz, Z., and Halter, W., 2009, Copper transport by high temperature, sul-
inclusions in quartz from northern Sonora: Implications for ore-genesis fur-rich magmatic vapor: Evidence from silicate melt and vapor inclusions
theory: Geology, v. 9, p. 3740. in a basaltic andesite from the Villarrica volcano (Chile): Earth and Plane-
Seedorff, E., Dilles, J.H., Proffett, J.M., Jr., Einaudi, M.T., Zurcher, L., tary Science Letters, v. 282, p. 115121.
Stavast, W.J.A., Johnson, D.A., and Barton, M.D., 2005, Porphyry deposits: Zajacz, Z., Hanley, J.J., Heinrich, C.A., Halter, W.E., and Guillong, M., 2009,
Characteristics and origin of hypogene features: ECONOMIC GEOLOGY 100TH Diffusive reequilibration of quartz-hosted silicate melt and fluid inclusions:
ANNIVERSARY VOLUME, p. 251298. Are all metal concentrations unmodified? [ext. abs.]: ECROFI Meeting,
Seo, J.H., Guillong, M., and Heinrich, C.A., 2009, The role of sulfur in the 20th, University of Granada, Spain, September 2009, Extended Abstracts,
formation of magmatic-hydrothermal copper-gold deposits: Earth and p. 285286.
Planetary Science Letters, in press. Zezin, D., Migdisov, A.A., and Williams-Jones, A.E., 2008, The solubility of
Sillitoe, R.H., 1997, Characteristics and controls of the largest porphyry cop- Au in H2O-H2S gas mixtures: Geochimica et Cosmochimica Acta, v. 72, p.
per-gold and epithermal gold deposits in the circum-Pacific region: Aus- A1076.
tralian Journal of Earth Sciences, v. 44, p. 373388.
2010, Porphyry copper systems: ECONOMIC GEOLOGY, v. 105, p. 341.
0361-0128/98/000/000-00 $6.00 118

The Bingham Canyon Porphyry Cu-Mo-Au Deposit. III. Zoned Copper-Gold Ore Deposition by Magmatic Vapor Expansion °

Uploaded by

Copyright:

Available Formats

The Bingham Canyon Porphyry Cu-Mo-Au Deposit. III. Zoned Copper-Gold Ore Deposition by Magmatic Vapor Expansion °

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Bingham Canyon Porphyry Cu-Mo-Au Deposit. III. Zoned Copper-Gold Ore Deposition by Magmatic Vapor Expansion °

Uploaded by

Copyright:

Available Formats

2010 Society of Economic Geologists, Inc.

Economic Geology, v. 105, pp. 91118

The Bingham Canyon Porphyry Cu-Mo-Au Deposit.

Introduction quantitative, integrating the role of magmatic fluids exsolved

Hybrid quartz monzonite porphyry 0.30 ppm Au grade

Porphyritic quartz monzonite 0.35 wt.-% Cu grade

SEDIMENTARY ROCK TYPES

which is a limitation of the present study. However, a few

TABLE 1. Summary Description of Samples, Veins, and Fluid Inclusions

Main Vein minerals Vein

Deep QMP, Potassic A51-4788 Molybdenite veinlet reopening a Kspar 8 QMP

A57-5910 Quartz molybdenite vein (fine- moly 8 QLP

D211-221 Quartz stockwork vein (vuggy, cpy, bn 8 QMP

Au-Cu QMP, Potassic D123-224 Quartz stockwork vein cpy 10 QMP

D211-176 Quartz-molybdenite vein (sparry, moly 10 QMP

4990-2050 Quartz-chalcopyrite-molybdenite moly 10 QMP

5090-1510 Quartz stockwork vein bn, dig, cpy 8 QMP

Observations, Ordered by Sampling Regions shown in Figure 3

Fluid inclusions: types, description and data extent

Sulfide Fluid # assem. # single FI # assem. # single FI

Potassic, weak SA/CL F/R ID,(B,V) ID: small inclusions,

Potassic/dissem- LA,MT,CL L 1 B,V,(A) B, V: regularly to negative B5 B59 B3 B8

Potassic, weak SA F B,V,(A) B, V: up to 45 m, irregularly

Potassic/minor SA/CL F/L 2 2 B,V,(A) B, V: up to 50 m; B: 1-2

Potassic/minor LA,MT,CL L 1 1 B,V,(A) B, V: multiple gen., B2/V5/A2 B26/ B3/V1 B5/V4

Potassic/dissem- MT,CL R 2 2 B,V,(A) B, V: up to 50 m, two gen., B1 B9

Potassic/minor LA,MT L B,V,(A) B, V: multiple gen., B3/ B44/ B2/V1 B9/V5

Potassic/dissem- MT,CL R B,V,(A) B, V: irregularly to B1 B13

Main Vein minerals Vein

Cu-rich EM Potassic D74-1719 Quartz stockwork vein (zoned, py, minor 15 EM

D169-582 Quartz stockwork vein cpy and cc 7 EM

D188-1277 Quartz stockwork vein (milky to py, cpy as 7 EM

D411-2646.5 Quartz-sericite-pyrite vein (py- py, cpy , bn, sericite 5 EM

D420-5192 Quartz stockwork vein (clear to py, minor cpy bio 5 EM

D430-3657 Two crosscutting quartz stockwork py, cpy 3/4 EM

D430-3749 Reopened quartz stockwork vein cpy, cc 5 EM

Fluid inclusions: types, description and data extent

Sulfide Fluid # assem. # single FI # assem. # single FI

Potassic/dissem- MT R B,V,(A) B, V: two gen., second gen. B1 B14

Potassic/dissem- LA,MT L B,V,(A) B, V: up to 15 m, some B1/V2 B10/V13 B2/V1 B8/V3

Potassic, propylitic LA,MT F B,V,(A) B, V: up to 50 m; B: 1-2 B2/V8 B13/V36 B2/V4 B11/V21

Potassic/dissem- SA/CL F/L 2 2 B,V,ID B, V: predominating, up to

Potassic, phyllic LA,MT,CL L 3 1 py A A: CO2 bearing, <10 m, A19 A148 A5 A10

Potassic, propylitic SA/CL F/L 1 1 ID,(B,V) ID: predominating, up to

Fluid inclusion types: ID = intermediate density; V = vapor; B = brine; A = aequeous liquid

quartz vein tendency for overestimating the average concentration ratio

0361-0128/98/000/000-00 $6.00 100

0361-0128/98/000/000-00 $6.00 101

0361-0128/98/000/000-00 $6.00 102

0361-0128/98/000/000-00 $6.00 103

s D188-1277 Cu-rich periphery l

0361-0128/98/000/000-00 $6.00 104

10 1 wt percent NaCl equiv (Furrer, 2006). Intermediate-den-

? 8 Center ple A57-6063), but numerous other intermediate-density as-

Cu-rich periphery 300 bar

100 bar Brine

0361-0128/98/000/000-00 $6.00 105

0361-0128/98/000/000-00 $6.00 106

have a somewhat lower dP/dT slope. Using the SoWat iso-