38 - Ioannou Spooner Barrie 2009 Matagami FL Incl 2007

©2007 Society of Economic Geologists, Inc.
Economic Geology, v. 102, pp. 691–715
Fluid Temperature and Salinity Characteristics of the

Matagami Volcanogenic Massive Sulfide District, Quebec
S. E. IOANNOU,†,* E. T. C. SPOONER,
F. G. Smith Fluid Inclusion Laboratory, Department of Geology, University of Toronto,
22 Russell Street, Toronto, Ontario, Canada M5S 3B1
AND C. T. BARRIE
C. T. Barrie and Associates, 29 Euclid Avenue, Ottawa, Ontario, Canada K1S 2W2
Abstract
Fluid inclusions hosted within various lithologic units of the >40 million metric ton (Mt) Matagami district,
Abitibi greenstone belt, preserve samples of Archean volcanogenic massive sulfide (VMS) and postvol-
canogenic massive sulfide hydrothermal fluid. Microthermometric measurements on ore-hosted primary two-
phase liquid-vapor inclusions from Matagami’s south limb deposits indicate that the VMS hydrothermal fluid
was highly saline (16.2 ± 4.7 wt % NaCl-CaCl2 equiv, 1σ, n = 230) and of moderate temperature (trapping tem-
perature, Tt = 208° ± 32°C, 1σ, n = 230). A fluid with these characteristics is capable of transporting ~5 × 10–4
m (30 ppm) Zn as ZnCl–3 and ZnCl2– 4 chloride complexes. However, the low temperature of this fluid precluded
efficient Cu transport (≤3 ppm), which may in part explain the relatively Cu poor nature of the Matagami de-
posits. Calculated densities of this ore fluid as high as 1.10 g/cm3 are consistent with a bottom-hugging brine
model. However, a subset of the data indicate a fluid less dense than ambient seawater, suggesting that buoy-
ant hydrothermal plumes were also present. A microthermometrically determined high CaCl2 content (XNaCl
<0.55, molar Na/Ca = 2.3/1) for the VMS ore-hosted primary fluid is consistent with an Archean hydrothermal
fluid more Ca-rich than modern-day seawater.
Quartz-epidote veins located in the hydrothermal cracking zone of the Bell River Complex host primary liq-
uid-vapor-halite inclusions. These inclusions are interpreted to be samples of the deep-seated equivalent to the
VMS ore-hosted hydrothermal fluid described above. Microthermometry indicates that these inclusions
trapped a high-temperature brine (Tt = 373° ± 44°C, 1σ, n = 92; 38.2 ± 1.9 wt % NaCl equiv, 1σ, n = 92). We
interpret this brine to be a phase-separated product of (modified) model seawater (3.2 wt % NaCl), an exsolved
magmatic fluid, or a combination thereof, deep within the hydrothermal system at 650° to 670°C and a near-
lithostatic pressure of 90 MPa. Phase separation and subsequent convection lowered the temperature of the
brine prior to its entrapment within the hydrothermal cracking zone.
The occurrence of high-temperature brine overlain by lower temperature/salinity fluid suggests a two-cell
convection model, consistent with metal mass-balance calculations for the south limb. The high salinity of the
ore-hosted fluid inclusions indicates two possibilities: (1) a significant amount of brine was incorporated into
the upper cell and mixed with heated seawater during convection; (2) Archean seawater itself was very saline
and of variable salinity.
With the cooling of the Bell River Complex, lower temperature fluids, dominantly of seawater origin, circu-
lated deep within the hydrothermal system. Modified by water-rock interaction, yet not phase separated, these
fluids sealed off the remnant permeability of the fracture network of the hydrothermal cracking zone and were
locally trapped as secondary liquid-vapor fluid inclusions (homogenization temperature, Th = 242° ± 17°C; 9.1
± 1.6 wt % NaCl equiv, 1σ, n = 14) hosted within the Bell River Complex quartz-epidote vein material.
Post- and/or waning-stage VMS hydrothermal activity is evident from the presence of quartz-epidote veins
crosscutting Wabassee Group hanging-wall rocks. Microthermometry on quartz-hosted primary liquid-vapor
fluid inclusions suggests that this activity occurred at relatively low temperatures (Th = 76°–177°C, n = 212),
over a wide range of salinity (6.0–32.4 wt % NaCl-CaCl2 equiv, n = 212), and with a high apparent CaCl2 con-
tent (XNaCl <0.06).
These fluid inclusion data illustrate the importance of subsea-floor chemical and physical processes directly
related to metal transport and deposition in VMS systems. In particular, phase separation deep within the hy-
drothermal system is interpreted as a key process for generating saline brines capable of forming significant ore
deposits.
Introduction the sea floor. Detailed research has now been carried out at
THE DISCOVERY by Francheteau et al. (1979) of metal-rich many such sites, including direct sampling and analysis of
black-smoker chimneys on the East Pacific Rise was the first venting hydrothermal fluids. Direct analyses of the fluids
of many subsequent discoveries of hydrothermal sulfides on have provided great insight into their chemical composition,
including the identification of parental end-member fluids,
† Corresponding author: e-mail, [email protected]
metal complexing, and transport-related characteristics (as
*Current Address: Haywood Securities Inc., 181 Bay Street, Suite 2910, reviewed by Scott, 1997; Seward and Barnes, 1997). It is gen-
Toronto, Ontario, Canada M5J 2T3. erally accepted that black-smoker sea-floor activity represents
0361-0128/07/3674/691-25 691
692 IOANNOU ET AL.
a modern analog of the hydrothermal systems responsible for and have ages that overlap within low uncertainties (+2.5/-1.9
producing ancient VMS deposits in the geologic record. MaU-Pb: Mortensen 1993a). In addition, geochemical similari-
Despite the economic significance of VMS deposits, rela- ties to the Watson Lake and Wabassee Group tholeiitic basalts
tively little has been done to quantify directly the chemical suggest that these lithologic units were derived through frac-
composition and physical properties of the hydrothermal flu- tionation processes associated with the Bell River Complex
ids responsible for ore deposition in ancient systems. A rea- (MacGeehan and MacLean, 1980a, b; Scott, 1980). A rela-
son for this lack of research is that deformation and/or meta- tively continuous Mg-Fe fractionation trend across the Bell
morphism can overprint the deposits, eliminating trapped River Complex suggests that it was formed primarily through
fluid inclusions of primary origin. one large continuous intrusive event (Maier et al., 1996).
Characterizing hydrothermal fluid chemistry is key to un- However, gabbro sills cut the VMS deposits, Watson Lake
derstanding metal transport in hydrothermal systems, and Group, and Wabassee Group rocks, indicating that less volu-
fluid inclusion microthermometric techniques can constrain a minous Bell River Complex-like magmas (based on field rela-
number of chemical and physical parameters, including the tionships and similar REE patterns) were emplaced after
temperature and bulk salinity of the fluid. In addition, such VMS formation and (early) deposition of the Wabassee Group
techniques can provide significant insights into fluid-mixing (Scott, 1980; Maier et al., 1996). Multiphase intrusions spa-
pathways, phase-separation processes, and variations in end- tially associated with VMS systems, which include pre- to syn-
member fluid compositions, all of which have significant im- and post-VMS phases, are not uncommon (e.g., Galley et al.,
pact on the ability of a fluid to transport and deposit metals. 2000a, b). At Matagami, available data suggest that a bulk of
Herein we present fluid inclusion data from the Archean the Bell River Complex was emplaced prior to and/or during
(2.72 Ga) Matagami VMS district in the Superior province, VMS formation.
Abitibi greenstone belt, of Quebec. In addition to providing Regional deformation, synchronous with and postdating
key information about the chemical and physical properties of the Bell River Complex intrusion and VMS mineralization,
the hydrothermal fluid responsible for VMS deposition in the formed the westward-plunging district-scale Galinée anti-
district, the study provides new insights into some of the in- cline (Piché et al., 1993). Subsequent erosion has exposed the
tricacies of VMS systems in the Archean, which are valuable Bell River Complex along the axis of the anticline and two
for both research and mineral exploration. volcanic belts on the north and south limbs (Fig. 1). Younger
banded-iron formations and clastic sediments, primarily silt-
Geologic Setting stones, argillites, and minor granitoid pebble conglomerates
Located in the Harricana-Turgeon belt of the northern of the Matagami and Taibi Groups, conformably bound these
Abitibi subprovince, Quebec (Lacroix et al., 1990), the limbs to the north and south, respectively (Beaudry and
Matagami mining camp is one of the largest multideposit Gaucher, 1986).
VMS districts in the Abitibi greenstone belt, containing in ex- The north-limb volcanic rocks dip steeply to the north and
cess of 40 Mt of massive sulfide ore, second only to Rouyn- are locally overturned. Coinciding with the Lac Garon defor-
Noranda (~103 Mt excluding the Horne 5 zone: Kerr and mation zone (Lacroix et al., 1990), the rocks are moderately
Gibson, 1993). Detailed descriptions of the geology of the to strongly deformed and sheared, creating lozenge-shaped
Matagami district can be found in Sharpe (1968), Roberts structural domains (Piché et al., 1993). Stratigraphic relation-
(1975), and Beaudry and Gaucher (1986). In brief, the stratig- ships are obscured (Sharpe, 1965), and amphibolite-grade
raphy of the district consists of a lower, dominantly felsic vol- metamorphic facies exist throughout (Jolly, 1978). In contrast,
canic package, the Watson Lake Group (> 1 km thick, 2724.5 the south-limb volcanic rocks are much less deformed, con-
± 1.8 MaU-Pb: Mortensen, 1993a), overlain by a dominantly sistently dipping 40° to 60° to the south (Piché et al., 1993).
mafic volcanic package: the Wabassee Group (>13 km thick). Stratigraphic relationships are readily apparent, the Key
A thin (2–4 m thick) unit of pyritic laminated cherty tuff Tuffite is continuous throughout, and primary textures are
known as the Key Tuffite marks the division between the two well preserved with only local overprinting by prehnite-
groups (Davidson, 1977; Liaghat and MacLean, 1992). The pumpellyite to greenschist-facies metamorphism (Sharpe,
unit is a chemical exhalative sedimentary rock averaging 1.4 1968; Ioannou et al., 2004).
wt percent Zn and 0.1 wt percent Cu (Davidson, 1977) that Postvolcanic to post-tectonic intrusions of varying composi-
has been interpreted to be genetically related to the VMS de- tions (dioritic, tonalitic, granodioritic) occur throughout the
posits of the district, which also occur, for the most part, at the district and include the Olga Lake (2693 ± 1.6 MaU-Pb, ~32
contact between the Watson Lake and Wabassee Groups. Ma younger: Mortensen, 1993a), Desmazures, Cavelier, and
The Bell River Complex, originally described by Freeman Dunlop Bay plutons. In addition, Proterozoic diabase dikes
(1939), is a >5,000-m-thick layered tholeiitic gabbro and/or intrude the stratigraphy throughout the western part of the
anorthosite body (2724.6 +2.5/–1.9 MaU-Pb: Mortensen, 1993a) camp (Piché et al., 1993).
that intrudes the lower Watson Lake Group. Many, including Twenty VMS deposits with tonnages ranging from ~0.1 to
Piché et al. (1990), Maier et al. (1996), and Ioannou and more than 25 Mt have been discovered in the Matagami dis-
Spooner (2007), have suggested that the Bell River Complex trict since 1956. The deposits are typically Zn rich, averaging
acted as the heat source responsible for driving Archean VMS- ~9 wt percent Zn and reaching as high as 17.6 wt percent
related hydrothermal circulation in the area. Supporting Zn (e.g., Isle-Dieu). Average Cu grades for the deposits are
evidence includes the fact that granophyre of the Bell River significantly lower, ~1 wt percent Cu, and do not exceed 2 wt
Complex, found in the uppermost level of the intrusion, and percent Cu. A number of characteristics are common to most
the Watson Lake Group rhyolite are compositionally identical deposits: (1) strata-bound disposition of the massive ore; (2)
0361-0128/98/000/000-00 $6.00 692

MATAGAMI VMS DISTRICT, ABITIBI, QUEBEC: FLUID INCLUSION CHARACTERISTICS 693
Dunlop
Matagami Lake Bay
NH
undar y Fault
Allard Nor ther n Bo
RW
River NO
N
NE 90 o GL
BC
G2
Garon
PW R2 Lake
Matagam i
EQ
PR
Olga
ID Lake
Watson
Lake Shallow
ML Be
ll R Lake
OW o
45 ive
r
OR
Lithology BA Gal
in e
Taibi Group eA
ntic
Sedimentar y Rocks lin e
BS
Intrusive Rocks
Diabase Dikes
Dunlop Bay Pluton (granitoid)
Olga Lake Pluton (granitoid)
Peridotite
Wabassee Group Da
Basalt n i el
Fa ?
Watson Lake Group ult
Rhyolite
Dacite to Basalt
Bell River Complex 0 km 5
Gabbro Sills
Granophyric Zone
Layered Zone (gabbro/anor thosite)
Main Zone (gabbro)
FIG. 1. Geologic map of the Matagami district (modified from Piché et al., 1993; Maier et al., 1996). North limb deposits:
BC = Bell Channel, GL = Garon Lake, G2 = Garon Lake 2, NE = Norita East, NH = New Hosco, NO = Norita, RW = Ra-
diore West, RS = Radiore 2. South limb deposits: BA = Bell Allard, BS = Bell Allard South, EQ = Equinox, ID = Isle-Dieu,
ML = Mattagami Lake, OR = Orchan, OW = Orchan West, PR = Perseverance, PW = Perseverance West.
well-defined upper contacts, with lower contacts gradational Roberts and Reardon, 1978; Costa et al., 1983; Ioannou,
through a zone of disseminated sulfides and highly altered foot- 2004). Unfortunately, this mineral assemblage is a poor host
wall volcanic rocks; (3) predominance of pyrite plus pyrrhotite, for fluid inclusions. Furthermore, quartz-sulfide stringer
with lesser amounts of sphalerite, magnetite, and chalcopyrite; veins, commonly found in VMS alteration pipes (e.g.,
and (4) on the south limb, a strong stratigraphic control on sul- Pisutha-Arnond and Ohmoto, 1983) are rare at Matagami,
fide mineralization, with most orebodies located at (within) the negating a fluid inclusion study of footwall alteration pipe
top of the Watson Lake Group, typically overlain by the Key material.
Tuffite (Sharpe, 1965; Roberts, 1975; Costa et al., 1983).
VMS ore material
Sampling Methodology As discussed above, the rocks of the south limb, including
Sampling for fluid inclusion-bearing material focussed on the contained ore deposits, preserve primary textures and
specific domains of the seven different VMS deposits in order have been metamorphosed to only prehnite-pumpellyite to
to characterize and compare the hydrothermal fluids greenschist-facies metamorphism overprinting locally, mak-
throughout the Matagami district. These domains include ing the south limb an ideal location to obtain primary VMS
VMS-ore (mound) material and deep footwall intrusive mate- ore fluids. Samples were obtained from Noranda Ltd.’s explo-
rial from the Bell River Complex. In addition, post- and/or ration drill core archive. The samples included massive sul-
waning-stage VMS veins were also collected from the hanging fide material from four deposits: Bell Allard (3.2 Mt), Isle-
wall of several deposits to characterize younger fluids in the Dieu (2.7 Mt), Orchan West (1.0 Mt), and Mattagami Lake
Matagami district. (25.6 Mt; Fig. 1).
VMS alteration pipes beneath the Matagami orebodies Sphalerite was targeted for fluid inclusion research for a
are characterized by complete alteration of Watson Lake number of reasons: (1) low Fe sphalerite can contain useable
Group rhyolites to chlorite ± talc (MacGeehan, 1978; inclusions and transmit visible light, allowing the use of
0361-0128/98/000/000-00 $6.00 693

694 IOANNOU ET AL.
standard microthermometric techniques (e.g., Costa et al., et al., 2004). Similar quartz-sulfide relationships have been
1983); (2) the Matagami hydrothermal system was extremely documented in modern sea-floor hydrothermal systems (e.g.,
Zn rich, and the orebodies are mined primarily for their high TAG: Knott et al., 1998).
Zn content (e.g., Isle-Dieu avg grade of 17.6% Zn); (3) quartz,
a mineral capable of preserving primary fluid inclusions Footwall intrusive material
through low- to middle-grade metamorphism, is commonly Abundant fractures and veins crosscutting the western
found associated with sphalerite. Detailed ore petrography lobe of the Bell River Complex contain a range of mineral as-
conducted on sphalerite-rich (> 60% ZnS) massive sulfide semblages representing a spectrum of temperatures from
samples indicates that the fluid inclusion-bearing quartz in 250° to 700°C. The 250° to 400°C assemblage, quartz-epi-
these samples was coprecipitated with the sphalerite; the dote ± sericite ± chlorite ± plagioclase (Kristmannsdóttir,
quartz is intimately intergrown with the sphalerite, occurring 1979; Richardson et al., 1987; Schiffman and Smith, 1988;
both as inclusions in the sphalerite and as host to sphalerite Harper, 1999; Gibson et al., 2000), is the most widespread
inclusions (Fig. 2A). Costa et al. (1983) noted a similar rela- and occurs as parallel, orthogonal, anastomosing, and ran-
tionship in massive sulfide samples from the Mattagami Lake dom vein sets (Fig. 2B). Vein densities average between 15
mine, and their comparable fluid inclusion homogenization and 25 veins per m2, locally reaching as high as 40 to 60 veins
temperatures and isotope fractionation-derived temperatures per m2. These veins range from 1 to 60 mm wide, averaging
derived from quartz and sphalerite support this association. 1 to 3 mm, and are interpreted to represent thermal cracking
Furthermore hot cathode cathodoluminescence of VMS ore (Lister, 1974, 1975) associated with hydrothermal fluid flow
material reveals primary concentric growth zones within the during VMS deposit formation in the district (Ioannou and
quartz that predates pulses of sulfide deposition and produces Spooner, 2007). The fineness of these veins is consistent with
a transient cathodoluminescence response characteristic of veins found in the hydrothermal cracking zone in plutonic
hydrothermal quartz—clear evidence that the quartz is unde- rocks in ophiolite settings (Nehlig, 1991, 1993). The veins
formed and directly related to VMS mineralization (Ioannou provide an opportunity to sample fluids from the deepest
(A) (B)
0 cm 1
Sph
Qtz
(C) (D)
FIG. 2. (A). Polished thin section of massive sulfide ore from the Bell Allard mine. Note the intergrown texture between
quartz and sphalerite interpreted to indicate coprecipitation (Qtz = quartz, Sph = sphalerite). (B). Outcrop of quartz-epidote
veins in the hydrothermal cracking zone of the Bell River Complex in the town of Matagami. (C). Drill core (NQ size) in-
tercept of a hanging-wall quartz-epidote vein near the Bell Allard mine (white = quartz, gray = epidote). (D). Outcrop of
quartz veins related to the Olga Lake pluton crosscutting the Bell River Complex near Shallow Lake.
0361-0128/98/000/000-00 $6.00 694

levels of the VMS system, ±1,650 m below the paleosea floor ice-melting determinations and ± 5°C for liquid-vapor ho-
(Ioannou and Spooner, 2007). The assemblage quartz-epi- mogenization and halite-melting determinations. Doubly pol-
dote is well suited for fluid inclusion research, as both min- ished (~60 µm) thin sections were prepared for each sample.
erals are capable of preserving primary hydrothermal fluid Primary and secondary fluid inclusions were identified using
inclusions even under low- to medium-grade metamorphic the criteria of Roedder (1984). Rare irregular and/or flattened
conditions. Furthermore, the transient cathodoluminescence fluid inclusions, suggestive of necking, were avoided. Tem-
response of quartz from these veins demonstrates a hy- perature cycling, as described by Haynes (1985), was used to
drothermal (i.e., nonrecrystallized) association (Ioannou et ensure accurate final solid-phase melting temperatures. The
al., 2004). relatively small (<20 µm) inclusion size in all sample types
precluded the distinction between ice and salt hydrates dur-
Late fluid-bearing material ing microthermometric freezing runs.
Although VMS ore formation had mainly ended by the time Supercooling the inclusions to –50° to –60°C in most sam-
of the Wabassee Group was deposited, continued waning- ple subsets was required to cause complete freezing. Upon
stage hydrothermal circulation is evident from the presence subsequent heating, minimum eutectic melting temperatures
of multiple exhalative tuffite layers nested within the (Te(ice)) measured in two-phase liquid-vapor VMS ore mater-
Wabassee volcanic rocks as well as from hanging-wall oxygen ial and hanging-wall quartz-epidote vein material, ranging
isotope (Carr, 2004) and mineral alteration (Ioannou, 2004) from –38° to –51°C, indicate the presence of salt species in
data. In order to characterize these and/or later fluids, a num- addition to NaCl, likely divalent cation chlorides (Te(MgCl2)=
ber of quartz-epidote (±chlorite) veins crosscutting the lower –35°C, Dubois and Marignac, 1997; Te(CaCl2) = –52°C, Davis
2 km of the Wabassee Group were sampled in drill core. Con- et al., 1990). Therefore, final ice-melting temperature (Tm(ice))
taining coarse-grained euhedral epidote rosettes intergrown values were converted into salinity using the NaCl-CaCl2-
with quartz (Fig. 2C), these veins are an excellent host for pri- H2O system data of Oakes et al. (1990). In secondary inclu-
mary fluid inclusions. The elevated temperature (72°–182°C; sions in quartz-epidote vein material from the Bell River
see below) and similar mineralogy of these hanging-wall veins Complex and in two-phase liquid-vapor primary inclusions in
to the Bell River Complex-hosted footwall veins described vein material related to the Olga Lake pluton, minimum mea-
above, as well as epidote patches and veins from gabbro dikes sured Te(ice) of –14°C indicates that the salinity of the inclu-
along the northern margin of the Bell River Complex near the sions is dominated by NaCl (Te(NaCl) = –21.2°C: Hall et al.,
Radiore 2 mine (Fig. 1), interpreted by Harrigan and 1988). Therefore, Tm(ice) values were converted to salinity
MacLean (1976) to represent channelways for hydrothermal using data for the NaCl-H2O system from Bodnar (1993).
fluids, suggest that they may be late derivatives of VMS fluids Application of the data from Oakes et al. (1990) requires an
or related to VMS-type fluids. The transient cathodolumines- estimate of the weight ratio (XNaCl) between NaCl and CaCl2.
cence response of quartz from these hanging-wall veins Vanko et al. (1988) provided a method for determining the
demonstrates this hydrothermal association (Ioannou et al., bulk composition of NaCl-CaCl2-H2O fluids, but it requires
2004). halite dissolution temperatures (Tm(salt)). Neither VMS ore
After the formation of the VMS deposits, the Matagami dis- material nor hanging-wall quartz-epidote vein material con-
trict underwent several episodes of granitoid plutonism in- tain inclusions with visible halite daughter crystals at room
cluding intrusion of the Olga Lake pluton, which is now lo- temperature. Therefore, an estimate of the maximum XNaCl
cated approximately 15 km to the east of the south limb was used instead. This maximum was determined by plotting
deposits (Fig. 1). Such episodes likely would have caused sig- the minimum Tm(ice) on the ternary NaCl-CaCl2-H2O plot of
nificant hydrothermal fluid flow in the subsurface and could Borisenko (1977; Fig. 3). A tie line drawn from the H2O apex,
have released orthomagmatic fluids. An excellent exposure of through the intersection of the –23.3°C isotherm and the ice-
quartz ± feldspar veins (up to 0.25 m wide) crosscutting lay- hydrohalite boundary, to the NaCl-CaCl2 join gives the maxi-
ered gabbro of the Bell River Complex, 2 km northwest of mum XNaCl for the inclusion population.
Shallow Lake (Fig. 2D), was sampled to help characterize the It should be noted that salinities calculated using the data
composition of the granite-related fluids. The veins are inter- from Oakes et al. (1990) are not significantly affected by
preted to have formed during intrusion of the nearby (<2,000 changes in XNaCl for fluids with Tm(ice) between –10° and
m) Olga Lake granitoid pluton (2693.2 ± 1.6 MaU-Pb: –14°C (e.g., for a fluid with Tm(ice) of –13°C, equivalent to ~16
Mortensen, 1993a). wt % total salt, a change in XNaCl of 0.5 results in only a ~0.3
wt % change in salinity). However, fluids with Tm(ice) >–10° or
Experimental Methodology <–14°C are more sensitive to changes in XNaCl given the ex-
Standard heating and freezing fluid inclusion microther- ponential nature of the data of Oakes et al. (1990; e.g., for a
mometry techniques, as described by Roedder (1984) and fluid with Tm(ice) of –23°C, equivalent to ~23 wt % total salt, a
Shepherd et al. (1985), were applied using a Linkam THMS change in XNaCl of 0.5 results in a ~1.5 wt % change in salin-
600 heating-and-freezing stage (Shepherd, 1981) attached to ity).
an Olympus BX50 microscope equipped with 10× oculars Supercooling primary three-phase, liquid-solid-vapor in-
and 5 to 100× objectives. The stage was calibrated (refer to clusions in quartz-epidote vein material from the Bell River
methodology of Macdonald and Spooner, 1981) at –56.6°, 0°, Complex to below –130°C did not nucleate an ice or salt hy-
and 374.1°C, using calibration-quality synthetic fluid inclu- drate solid phase. Therefore, no freezing data was obtained
sions from Bubbles Inc. of Blacksburg, Virginia, United from this inclusion population, limiting the ability to estimate
States. Measurement accuracy is estimated at ±0.2°C for final salt composition in the NaCl-CaCl2-H2O model system
0361-0128/98/000/000-00 $6.00 695

696 IOANNOU ET AL.
wt.% H2O
100
o
-5 C
o
-10 C
Ice P)
+ a r tz (o
Qu 2 C
o
-15 C )
Liquid S Ore -23. t e (P
VM m mi n = p ido o C
o
-20 C E .4
T ns 8 )
Vei = -2 r tz (S
Hydrohalite + HW m mi n Qua 8o C
T re 7. P)
Liquid S O = -3 ar t z o(
M
V m mi n u C
T eins Q 50. 6
V = -
H W m mi n
T
Halite + Liquid CaCl2 6H2 O
+
Liquid
wt.% NaCl wt.% CaCl2
40 40
)
)
5 (P)
XN ins 9 S)
6 (P
0 (P
Ve 0 tz (
0.0 ar tz
0.5 r tz
0.3 dote
r
a
HWaCl = ua
Qu
Qu
i
Ep
Q
0 .
XN Ore
XN Ore
XN ins
l =
l =
l =
Ve
S
S
aC
aC
VM
VM
aC
HW
FIG. 3. Ternary plot for the system H2O-NaCl-CaCl2 (from Borisenko, 1977), illustrating the method used to determine
the maximum XNaCl for the various fluid populations of this study. P = primary, S = secondary.
(Williams-Jones and Samson, 1990). However, given the pres- for a significant number of inclusions (33), Tm(salt) > Th(L-V(L)),
ence of CaCl2 in the VMS ore-related inclusions, it is not un- indicating that these fluids were trapped under P-T condi-
reasonable to expect CaCl2 in the primary inclusions in tions such that their isochore intersected the liquidus before
quartz-epidote vein material from the Bell River Complex. intersecting the liquid-vapor curve during cooling. Upon
Roedder (1984) noted that systems involving NaCl and diva- heating, Tm(salt), which corresponds to total homogenization,
lent cations such as Ca are susceptible to the salting-out ef- occurred along the halite liquidus. Bodnar (1994) has shown
fect; the addition of the divalent cation greatly decreases the that the slope of the halite liquidus varies with salinity, switch-
solubility of NaCl. For instance, the data of Linke (1958) in- ing from negative below ~50 wt percent NaCl to positive
dicate a solubility (at saturation) of only 1.02 wt percent NaCl above. Therefore, using the data of Sterner et al. (1988) to
at 25°C in the H2O-NaCl-CaCl2 system. Clynne et al. (1981) calculate the salinity of inclusions that homogenize along the
determined the solubility of NaCl in CaCl2 solutions up to liquidus by halite dissolution can significantly overestimate
100°C and found it to vary widely. Nevertheless, given the (>50 wt % NaCl) or underestimate (<50 wt % NaCl) the true
current lack of experimental data regarding the solubility of salinity.
NaCl in the presence of divalent cations at elevated tempera- The accurate calculation of salinity for inclusions where
tures, and the absence of freezing data, the salinity of these Tm(salt) > Th(L-V(L)) requires the pressure at which an inclusion’s
halite-saturated inclusions has been modeled using the H2O- isochore intersects the liquidus (Pl). Unfortunately a lack of
NaCl system. data defining the slopes of isochores in the liquid + halite
During heating above 0.0°C, in a majority of the three- field, as a function of salinity, inhibits the determination of
phase inclusions measured, melting of the salt (Tm(salt)) oc- this pressure. However, Bodnar (1994) has approximated the
curred prior to homogenization of the liquid and vapor to the slopes of iso-Th lines in the liquid + halite field for a 40 wt
liquid phase (Th(L-V(L))). For these inclusions, Tm(salt), in con- percent NaCl solution. Given the salinity (38.5 ± 2.2 wt %
junction with data of Sterner et al. (1988) for the H2O-NaCl NaCl equiv) calculated for inclusions in quartz-epidote vein
system, was used to calculate salinity. The data of Sterner et material from the Bell River Complex, where Tm(salt) < Th(L-
al. (1988) is theoretically valid only for inclusions in which V(L)) (see below), the 40 wt percent NaCl liquid + halite iso-Th
Tm(salt) = Th(L-V(L)) but can be used to approximate salinity for line data of Bodnar (1994) was used to approximate the Pl and
inclusions in which Tm(salt) < Th(L-V(L)) (Chou, 1987). However, salinity of inclusions where Tm(salt) > Th(L-V(L)). Pl is calculated
0361-0128/98/000/000-00 $6.00 696

from Th(L-V(L)), Tm(salt), Bodnar’s (1992) equation for three- Previous Fluid Inclusion Work
phase curve and linearly regressed iso-Th slope values of Bod- Through the use of fluid inclusion microthermometry,
nar (1994; as a function of Th(L-V(L)) in the liquid + halite field. Costa (1984) and Costa et al. (1983, 1984) determined an ore
Salinity is calculated via Bodnar’s (1994) equation for the fluid salinity of 3.5 wt percent NaCl equiv for samples from
slope of the liquidus, Bodnar’s (1992) equation for the three- the Mattagami Lake mine. Their data includes four types of
phase curve, and our own computer algorithm (unpub.). The primary fluid inclusions from massive sulfide ore: (1) two-
algorithm is based on the fact that only one distinct salinity- phase liquid > vapor inclusions with Tm values of –3° to –7°C,
defined liquidus will pass through a given [Pl, Tm(salt)] point in corresponding to a salinity of 3 to 7 wt percent NaCl equiv, (2)
P-T space; given the equation of the liquidus line, the algo- two-phase liquid > vapor inclusions with Tm values of –28° to
rithm determines if the point lies on it or not. When the al- –30°C, corresponding to a salinity of 22 to 27 wt percent
gorithm finds the liquidus line that passes through the [Pl, NaCl equiv, (3) three-phase liquid-vapor-solid (halite) inclu-
Tm(salt)] point, the salinity is defined (refer to App.). sions with Th(L-V(L)) ≈ Tm(salt) values of 240° to 280°C, corre-
In nonboiling hydrothermal systems, such as Matagami, sponding to a salinity of 35 to 38 wt percent NaCl equiv, and
fluid inclusion homogenization temperature (Th) values are a (4) liquid-solid (halite) inclusions.
minimum estimate of the true trapping temperature (Tt) of a
given fluid and, therefore, require a pressure correction to
convert the measured Th value into a Tt value (Potter and Microthermometry Results
Brown, 1977; Roedder and Bodnar, 1980). In this study, Th A total of 36 samples were examined for fluid inclusions. Of
measurements of primary fluid inclusions in VMS ore were these, 21 were found to contain measurable fluid inclusions
converted into Tt values using the multicomponent NaCl- from which 647 Th and Tm pairs were measured. The samples
CaCl2-H2O system data of Zhang and Frantz (1987) in con- can be divided into four groups: VMS ore material, quartz-
junction with an estimated fluid-trapping pressure (Pt) of 35 epidote vein material from the Bell River Complex, hanging-
MPa. This pressure is based on a 3,500-m estimated water wall quartz-epidote vein material, and vein material related to
depth during the time of VMS mineralization (MacGeehan, the Olga Lake pluton (Table 1). The predominant inclusion
1979, supported by data from MacGeehan, 1978). Note that habit in all sample types investigated is equidimensional to
the resultant Tt is not adversely affected by large changes in slightly elongate, similar to the oblate morphology described
estimated water depth. For example, a 1,000-m decrease in by Shepherd et al. (1985).
water depth will decrease the Tt temperatures by only ~10°C.
The Th((L-V) ± H(L ± H)) values of primary three-phase inclusions VMS ore material
in quartz-epidote vein material from the Bell River Complex The 23 samples of massive sulfide ore examined in this
were converted into Tt values using a Pt of 90 MPa. This pres- study consist primarily of a quartz-sphalerite assemblage as
sure assumes lithostatic conditions within the hydrothermal described above; 11 of these samples contained measurable
cracking zone, based on a 3,500-m estimated water depth dur- fluid inclusions. Fluid inclusion microthermometry produced
ing the time of VMS mineralization (MacGeehan, 1979; ρ = 1.0 230 primary Th(L-V(L))-Tm(ice) pairs as well as 43 secondary Th(L-
g/cm3), a Watson Lake Group stratigraphic thickness of 1,000 V(L))-Tm(ice) pairs, all of which came from quartz-hosted inclu-
m (Roberts, 1975; ρ = 2.6 g/cm3), and a stratigraphic thickness sions; no measurable fluid inclusions were located in spha-
of 1,000 m to the base of the hydrothermal cracking zone lerite. Six of the samples are from Bell Allard, three from
(Ioannou and Spooner, 2007; ρ = 2.9 g/cm3). Isle-Dieu, one from Orchan West, and one from Mattagami
Pressure corrections for inclusions with Tm(salt) < Th(L-V(L)) Lake (Tables 2, 3).
utilized the NaCl-H2O system liquid-vapor curve data of Primary Inclusions: Primary fluid inclusions contain two
Bischoff (1991) to calculate the liquidus pressure (Pl). The phases at room temperature: liquid (~85 vol %) and vapor
slope of the fluid’s iso-Th line in the liquid field was obtained (~15 vol %); no daughter minerals are present (Fig. 4A). The
from the data of Bodnar and Vityk (1994). Pressure correc- measured inclusions typically range in size from 1 to 8 µm
tions for inclusions with Tm(salt) > Th(L-V(L)) utilized the NaCl- (long dimension; mean of ~3 µm). As summarized in Tables 1
H2O system data of Bodnar (1994) to calculate Pl (as outlined and 2 and Figures 5 and 6, Tm(ice) values (mean = –13.2° ±
above). The slope of the fluid’s iso-Th line in the liquid field 5.2°C, 1σ, n = 230) indicate a salinity of 16.2 ± 4.7 wt percent
was approximated from the 40 wt percent NaCl solution data NaCl -CaCl2 equiv (1σ, n = 230) for primary VMS ore fluids.
of Bodnar (1994). These Pl values, along with Pt subsequently Primary inclusions hosted in VMS ore material all homoge-
define the trapping temperature (Tt) of primary fluid inclu- nized into the liquid phase upon heating, yielding an average
sions hosted by quartz-epidote vein material from the Bell Th(L-V(L))of 191° ± 31°C (1σ, n = 230). A relatively small cal-
River Complex (refer to App.). The density of these fluids was culated pressure correction of ~17°C yields a relatively low
calculated by applying the Th and salinity data to a regression ore fluid temperature of 208° ± 32°C (1σ, n = 230). The mea-
of the data of Urusova (1975) for the NaCl-H2O system. sured Th and salinity values indicate a fluid density of 1.03 ±
Apart from the primary fluid inclusions in quartz-epidote 0.05 g/cm3 (1σ, n = 230) at 208° ± 32°C.
vein material from the Bell River Complex, the density of all Secondary Inclusions: Secondary fluid inclusions occur as
other fluids is based on Th and salinity values calculated planar groups defining healed fractures, locally cutting across
herein and applied to Zhang and Frantz’s (1987) density data boundaries of adjacent quartz grains. The measured inclu-
for the NaCl-CaCl2-H2O system, in conjunction with a cor- sions range in size from 1 to 6 µm (long dimension, mean of
rection for the thermal expansion of quartz where applicable ~3 µm; Fig. 4B) and contain two phases, liquid (~90 vol %)
(Skinner, 1966; Bodnar and Sterner, 1985). and vapor (~10 vol %) at room temperature. As summarized
0361-0128/98/000/000-00 $6.00 697

698 IOANNOU ET AL.
TABLE 1. Summary of Fluid Inclusion Data from the Matagami VMS ores, from the Bell River Complex, and the Olga Lake Pluton1
Inclusion host and Th(L-V ± H(L ± H) Density

Group inclusion type (°C) Tt (°C)2 Tm(ice/salt) (°C) Salinity (wt %)3 Salinity (m)3 (g/cm3)4
VMS ore Quartz 191 ± 31 208 ± 32 –13.2 ± 5.2 16.2 ± 4.7 2.7 ± 0.9 1.03 ± 0.05
(n = 11) Primary (146 to 267) (162 to 285) (–23.2 to –0.9) (1.8 to 23.3) (0.2 to 4.1) (0.92 to 1.10)
(n = 230) (n = 230) (n = 230) (n = 230) (n = 230) (n = 230)
VMS ore Quartz 114 ± 14 na –27.1 ± 4.4 23.9 ± 1.8 3.1 ± 0.3 1.16 ± 0.02
(n = 4) Secondary (87 to 142) (–37.8 to –19.5) (20.4 to 27.9) (2.5 to 3.8) (1.13 to 1.22)
(n = 43) (n = 43) (n = 43) (n = 43) (n = 43)
BRC veins Quartz 324 ± 46 373 ± 44 302 ± 23 38.2 ± 1.9 10.6 ± 0.9 1.06 ± 0.03
(n = 3) Primary (259 to 452) (302 to 491) 267 to 370) (35.6 to 44.1) (9.5 to 13.5) (0.96 to 1.11)
(n = 92) (n = 92) (n = 92) (n = 92) (n = 92) (n = 92)
BRC veins Quartz 242 ± 17 na –6.0 ± 1.2 9.1 ± 1.6 1.7 ± 0.3 0.90 ± 0.02
(n = 1) Secondary (215 to 271) (–7.9 to –4.0) (6.5 to 11.5) (1.2 to 2.2) (0.86 to 0.91)
(n = 14) (n = 14) (n = 14) (n = 14) (n = 14)
HW veins Quartz 124 ± 23 na –21.3 ± 11.7 20.1 ± 5.9 2.5 ± 0.9 1.12 ± 0.06
(n = 5) Primary (76 to 177) (–50.6 to –3.1) (6.0 to 32.4) (0.6 to 4.6) (0.99 to 1.27)
(n = 212) (n = 212) (n = 212) (n = 212) (n = 212)
HW veins Epidote 162 ± 9 na –9.8 ± 6.9 12.7 ± 5.6 1.7 ± 0.9 1.03 ± 0.05
(n = 32) (n = 32) (n = 32) (n = 32) (n = 32)
OL veins Quartz 340 ± 20 na –5.1 ± 0.4 8.0 ± 0.6 1.5 ± 0.1 0.73 ± 0.04
(n = 24) (n = 24) (n = 24) (n = 24) (n = 24)
1 Top number gives the mean (± corresponds to 1σ), bracketed numbers give the range, n = the number of measurements, T = temperature of homoge-
h
nization, Tt = temperature of trapping, Tm = temperature of final ice melting, BRC = Bell River Complex, VMS = volcanogenic massive sulfide, HW = hang-
ing wall, OL = Olga Lake
2 T calculated using a pressure correction of 35 MPa for the massive sulfide ore and 90 MPa for the BRC quartz-epidote vein material fluid inclusions;
t
these pressures are based on a water depth of 3,500 m (MacGeehan, 1979), a Watson Lake Rhyolite stratigraphic thickness of 1,000 m (Roberts, 1975), and
a BRC thickness above the hydrothermal cracking zone of 1,000 m (Ioannou and Spooner, 2007) during the time of mineralization
3 VMS ore and HW vein salinity (NaCl-CaCl equiv) determined using the NaCl-CaCl2-H2O system of Oakes et al. (1990); BRC primary salinity ( NaCl
2
equiv) determined using the data of Bodnar (1994; Tm(salt) > Th(L(–V ± H(L ± H)) and Sterner et al. (1988; Tm(salt) < Th(L(–V ± H(L ± H)); Ol vein
and BRC secondary vein salinity (NaCl equiv) determined using the data of Bodnar (1993)
4 VMS ore, HW vein, secondary BRC vein, and OL vein density determined using the data of Zhang and Frantz (1987); primary BRC vein density de-
termined using the data of Usurova (1975)
in Tables 1 and 3 and in Figures 5 and 6, Tm(ice) values (mean well as 14 secondary Th(L-V(L))-Tm(ice) pairs; all of which came
= –27.1° ± 4.4°C, 1σ, n = 43) indicate a salinity of 23.9 ± 1.8 from quartz-hosted inclusions.
wt percent NaCl-CaCl2 equiv (1σ, n = 43), significantly higher Primary Inclusions: Primary fluid inclusions contain three
than the average value for primary inclusions of 16.2 ± 4.7 wt phases at room temperature: liquid (~60–72 vol %), solid
percent NaCl-CaCl2 equiv but overlapping the upper part of (~20 vol %), and vapor (~8–20 vol %; Fig. 4C). The solid
this range. phase consists of a singular well-defined cubic isotropic crys-
Secondary fluid inclusions hosted in VMS material all ho- tal interpreted to be NaCl. The measured inclusions range in
mogenized into the liquid phase upon heating, yielding an av- size from 2 to 8 µm (long dimension, mean of ~3 µm). As dis-
erage Th(L-V(L))of 114° ± 14°C (1σ, n = 43), ~80°C lower than cussed above, these inclusions exhibit both Tm(salt) < Th(L-V(L))
that of the primary inclusions. Given the current lack of data and Tm(salt) > Th(L-V(L)) microthermometric behavior, which to-
to constrain the timing, and hence the original depth during gether indicate an average fluid salinity of 38.2 ± 1.9 wt per-
this pulse of secondary fluid, the data were not pressure cor- cent NaCl equiv (1σ, n = 92) and a Tt of 373° ± 44°C (1σ, n
rected. The Th and salinity values indicate that fluids trapped = 92), with an upper range of ~480°C (Table 1, Figs. 5, 6).
by the secondary inclusions have a density of 1.16 ± 0.02 The Th and salinity values measured indicate a fluid density of
g/cm3 (1σ, n = 43). 1.06 ± 0.03 g/cm3 (1σ, n = 92).
Secondary Inclusions: Secondary fluid inclusions occur as
Quartz-epidote vein material from the Bell River Complex planar groups outlining healed fractures, locally cutting
Seven samples of quartz-epidote vein material from the across boundaries of adjacent quartz grains. The measured
Bell River Complex were studied; three of these samples con- inclusions range in size from 1 to 5 µm (long dimension,
tained measurable fluid inclusions. Fluid inclusion mi- mean of ~3 µm) and contain two phases, liquid (~85 vol %)
crothermometry produced 92 primary Th(L-V(L))-Tm(ice) pairs as and vapor (~15 vol %) at room temperature. As summarized
0361-0128/98/000/000-00 $6.00 698

TABLE 2. Summary of Data for Primary Fluid Inclusions in Quartz from Matagami VMS Ore1
Salinity (wt % Density

Deposit Inclusion host Th(L-V(L)) (°C) Tt (°C)2 Tm(ice) (°C) NaCl-CaCl2 equiv)3 Salinity (m)3 (g/cm3)4
Bell Allard Quartz 195 ± 33 212 ± 34 –12.7 ± 5.1 15.8 ± 4.7 2.6 ± 0.9 1.02 ± 0.04
(n = 6) (146 to 267) (162 to 285) (–23.2 to –0.9) (1.8 to 23.3) (0.2 to 4.1) (0.92 to 1.10)
(n = 164) (n = 164) (n = 164) (n = 164) (n = 164) (n = 164)
Isle Dieu Quartz 174 ± 17 191 ± 18 –15.2 ± 5.6 17.7 ± 4.9 3.0 ± 0.9 1.05 ± 0.05
(n = 3) (146 to 220) (162 to 236) (–21.9 to –2.2) (4.1 to 22.6) (0.6 to 3.9) (0.92 to 1.10)
(n = 48) (n = 48) (n = 48) (n = 48) (n = 48) (n = 48)
Orchan West Quartz 162 ± 9 178 ± 10 –15.2 ± 3.6 18.2 ± 2.7 3.0 ± 0.5 1.07 ± 0.03
(n = 1) (149 to 174) (165 to 191) (–19.5 to –10.7) (14.7 to 21.2) (2.3 to 3.6) (1.03 to 1.10)
(n = 7) (n = 7) (n = 7) (n = 7) (n = 7) (n = 7)
Mattagami Lake Quartz 215 ± 10 233 ± 9 –11.9 ± 3.5 15.5 ± 3.0 2.5 ± 0.6 1.00 ± 0.02
(n = 1) (196 to 226) (215 to 243) (–18.3 to –6.8) (10.6 to 20.5) (1.6 to 3.5) (0.97 to 1.04)
(n = 11) (n = 11) (n = 11) (n = 11) (n = 11) (n = 11)
Mean Quartz 191 ± 31 208 ± 32 –13.2 ± 5.2 16.2 ± 4.7 2.7 ± 0.9 1.03 ± 0.05
(n = 11) (146 to 267) (162 to 285) (–23.2 to –0.9) (1.8 to 23.3) (0.2 to 4.1) (0.92 to 1.10)
(n = 230) (n = 230) (n = 230) (n = 230) (n = 230) (n = 230)
1 Top number gives the mean (± corresponds to 1σ), bracketed numbers gives the range, n = the number of measurements, T = temperature of homog-
h
enization, Tt = temperature of trapping, Tm = temperature of final ice melting
2 T calculated using a pressure correction of 35 MPa (based on a water depth of 3,500 m; MacGeehan, 1979)
t
3 Salinity determined using the NaCl-CaCl -H O system of Oakes et al. (1990; X
2 2 NaCl = 0.55)
4 Density determined using the data of Zhang and Frantz (1987)
TABLE 3. Summary of Data for Secondary Fluid Inclusions in Quartz from Matagami VMS Ore1
Salinity (wt % Density

Deposit Inclusion host ThL(-V(L)) (°C) Tt (°C)2 Tm(ice) (°C) NaCl-CaCl2 equiv)3 Salinity (m)3 (g/cm3)4
Bell Allard Quartz 116 ± 15 na -27.3 ± 4.7 23.9 ± 1.9 3.1 ± 0.3 1.16 ± 0.03
(n = 3) (87 to 142) (-37.8 to -19.5) (20.4 to 27.9) (2.5 to 3.8) (1.13 to 1.22)
(n = 37) (n = 37) (n = 37) (n = 37) (n = 37)
Isle Dieu Quartz 106 ± 4 na -25.7 ± 1.0 23.4 ± 0.4 3.0 ± 0.1 1.17 ± 0.01
(n = 1) (102 to 111) (-27.0 to -24.1) (22.7 to 24.0) (2.9 to 3.1) (1.16 to 1.17)
(n = 6) (n = 6) (n = 6) (n = 6) (n = 6)
Mean Quartz 114 ± 14 na -27.1 ± 4.4 23.9 ± 1.8 3.1 ± 0.3 1.16 ± 0.02
(n = 4) (87 to 142) (-37.8 to -19.5) (20.4 to 27.9) (2.5 to 3.8) (1.13 to 1.22)
(n = 43) (n = 43) (n = 43) (n = 43) (n = 43)
1 Top number gives the mean (± corresponds to 1σ), bracketed numbers gives the range, n = the number of measurements, T = temperature of homog-
h
enization, Tt = temperature of trapping, Tm = temperature of final ice melting
2 Secondary ore inclusions are not presure corrected
3 Salinity determined using the NaCl-CaCl -H O system of Oakes et al. (1990; X
2 2 NaCl = 0.09)
4 Density determined using the data of Zhang and Frantz (1987)
in Table 1 and Figure 6, the secondary inclusions in quartz- zone of the Bell River Complex a model Tt of 300° ± 20°C,
epidote vein material from the Bell River Complex are char- 1σ, n = 14, is obtained). The measured Th and salinity values
acterized by a salinity of 9.1 ± 1.6 wt percent NaCl equiv indicate a fluid density of 0.90 ± 0.02 g/cm3 (1σ, n = 14).
(1σ, n = 14).
The fluid inclusions all homogenized into the liquid Hanging-wall quartz-epidote vein material
phase upon heating, yielding an average Th(L-V(L)) of 242° ± Six samples analyzed consisted of quartz-epidote vein ma-
17°C (1σ, n = 14). Given the current lack of data to con- terial which crosscut hanging-wall Wabassee mafic volcanic
strain the timing and the geology of the surrounding envi- rocks; five of these samples contained measurable fluid inclu-
ronment during this pulse of secondary fluid activity, the sions. Fluid inclusion microthermometry produced 212 pri-
data were not pressure corrected and thus represent mini- mary Th(L-V(L))-Tm(ice) pairs from quartz-hosted inclusions, as
mum trapping temperatures (if the same pressure is used well as 32 primary Th(L-V(L))-Tm(ice) pairs from epidote-hosted
as for the primary inclusion in the hydrothermal cracking inclusions.
0361-0128/98/000/000-00 $6.00 699

700 IOANNOU ET AL.
summarized in Tables 1 and 2 and Figures 5 and 6, Tm(ice) val-

(A)
ues (mean = –21.3° ± 11.7°C, 1σ, n = 212) indicate quartz-
hosted fluids from quartz-epidote hanging-wall vein material
have a salinity of 20.1 ± 5.9 wt percent NaCl-CaCl2 equiv (1σ,
vapor n = 212).
liquid The quartz-hosted inclusions all homogenized into the liq-
uid phase upon heating, yielding an average Th(L-V(L)) of 124° ±
2 µm 23°C (1σ, n = 212). Although this vein set has been found
crosscutting the Wabassee volcanic rocks more than 2 km
(B) above the ore horizon, its actual paleodepth of formation re-
mains unknown. Therefore, the Th data were not pressure cor-
rected and thus represent minimum trapping temperatures.
The measured Th and salinity values indicate that the fluids re-
liquid + vapor
sponsible for quartz deposition in these hanging-wall veins had
a relatively high density of 1.12 ± 0.06 g/cm3 (1σ, n = 212).
Primary epidote-hosted inclusions: Primary fluid inclusions
15 µm hosted in epidote are similar in size and contain the same
phases as quartz-hosted inclusions described above. As sum-
(C) marized in Tables 1 and 2, and Figures 5 and 6, Tm(ice) values
(mean = –9.8° ± 6.9°C, 1σ, n = 32) indicate that epidote-
hosted primary fluid inclusions from quartz-epidote hanging-
wall vein material have a salinity of 12.7 ± 5.6 wt percent
liquid NaCl-CaCl2 equiv (1σ, n = 32), overlapping that of the
vapor
solid
quartz-hosted inclusions at the 1σ level.
The epidote-hosted inclusions all homogenized into the liq-
1 µm
uid phase upon heating, yielding an average Th(L-V(L)) of 162°
± 9°C (1σ, n = 32), slightly higher than that of the quartz-
(D) hosted inclusions at the 1σ level. As with the quartz-hosted
inclusions, the Th(L-V(L)) data for the inclusions in epidote were
liquid not pressure corrected. Based on data from active geothermal
vapor systems (e.g., Reyes, 1990), epidote has a minimum thermal
stability of about 250°C, which implies an average Pt of 180
MPa or a 6.7-km paleodepth of formation for the epidote-
4 µm hosted inclusions (assuming ρWabassee = 2.7 g/cm3; calculated
using the data of Zhang and Frantz, 1987). This depth is not
unreasonable given the estimated total thickness of 13 km of
(E)
the Wabasse Group (Sharpe, 1968). The measured Th and
salinity values indicate that the fluids responsible for epidote
deposition in these hanging-wall veins had a density of 1.03 ±
0.05 g/cm3 (1σ, n = 32), overlapping the quartz-hosted inclu-
vapor sion data at the 1σ level.
liquid
Vein material related to the Olga Lake pluton
3 µm
Two analyzed samples of white quartz vein material inter-
FIG. 4. Fluid inclusion photomicrographs (all micrographs are taken at preted to be related to the Olga Lake pluton both contained
room temperature). (A). Primary liquid-vapor fluid inclusion hosted in quartz
in VMS ore. (B). Plane of secondary liquid-vapor fluid inclusions hosted in
measurable primary fluid inclusions. Fluid inclusion mi-
quartz in VMS ore (secondary fluid inclusions in quartz-epidote veins from crothermometry produced 24 Th(L-V(L))-Tm(ice) pairs. The in-
the hydrothermal cracking zone of the Bell River Complex are similar in ap- clusions contain two phases at room temperature, liquid (~85
pearance). (C). Primary liquid-vapor-solid (halite) fluid inclusion in quartz in vol %) and vapor (~15 vol %); no daughter minerals are pre-
a quartz-epidote vein from the hydrothermal cracking zone of the Bell River sent (Fig. 4E). The measured inclusions range in size from 2
Complex. (D). Primary liquid-vapor fluid inclusions in quartz from a hang-
ing-wall quartz-epidote vein. (E). Primary liquid-vapor fluid inclusions in to 8 µm (long dimension, mean of ~4 µm). As summarized in
quartz from a vein related to the Olga Lake pluton. Tables 1 and 2 and Figures 5 and 6, Tm(ice) values (mean =
–5.1° ± 0.4°C, 1σ, n = 24) indicate that primary fluid inclu-
sions in quartz vein material related to the Olga Lake pluton
have a salinity of 8.0 ± 0.6 wt percent NaCl equiv (1σ, n = 24).
Primary quartz-hosted inclusions: Primary fluid inclusions The fluid inclusions all homogenized into the liquid phase
hosted in quartz contain two phases at room temperature, upon heating, yielding an average Th(L-V(L)) of 340° ± 20°C
liquid (~85 vol %) and vapor (~15 vol %); no daughter min- (1σ, n = 24). Given the current lack of data to constrain the
erals are present (Fig. 4D). The measured inclusions range geometry of the surrounding environment during intrusion of
in size from 3 to 18 µm (long dimension, mean of ~7 µm). As the Olga Lake pluton and the pulse of this associated fluid,
0361-0128/98/000/000-00 $6.00 700

(A) VMSprimary Tt Histogram (B) VMSprimary Salinity Histogram (C) VMS primary Density Histogram
80 120 60
60 90
Frequency
40
Frequency
Frequency
40 60
20
20 30
0 0 0
0
0
0
0
0
0
0
0
0
0
0
0.9
0.9
1.0
1.0
1.1
1.1
1.2
0
10
20
30
40
50
16
20
24
28
32
36
40
44
48
52
Tt (oC) Salinity (eq.wt.% NaCl-CaCl2) Density (g/cm3)
(D) BRC primary Tt Histogram (E) BRC primary Salinity Histogram (F) BRC primary Density Histogram
25 100 60
20 80
Frequency
40
Frequency
Frequency
15 60
10 40 20
5 20
0 0
0
0
0
0
0
0
0
0
0
0
0
0
0.9
0.9
1.0
1.0
1.1
1.1
1.2
16
20
24
28
32
36
40
44
48
52
10
20
30
40
50
Tt (oC) Salinity (eq.wt.% NaCl) Density (g/cm3)
FIG. 5. Histograms of microthermometric measurements of primary fluid inclusions in VMS ore and in quartz-epidote
veins from the Bell River Complex. (A). Trapping temperature (Tt) of fluid inclusions in quartz in VMS ore. (B). Salinity of
fluid inclusions in quartz in VMS ore. (C). Densities of fluid inclusions in quartz in VMS ore sulfide. (D). Trapping temper-
ature (Tt) of fluid inclusions in quartz-epidote veins of the Bell River Complex. (E). Salinity of fluid inclusions in quartz-epi-
dote veins of the Bell River Complex. (F). Densities of fluid inclusions in quartz-epidote veins of the Bell River Complex.
50
40
Salinity (wt.%)
30 VMS (primary: Tt)
VMS (secondary : Th)
20 BRC (primary: Tt)
BRC (secondary : Th)
HW (primary : Th)
10
OL (primary :Th)
0
0 100 200 300 400 500
Tt / Th ( oC)
FIG. 6. Salinity vs. Tt/Th plot for the Matagami fluids. BRC = Bell River Complex, HW = hanging wall, OL = Olga Lake,
VMS = volcanogenic massive sulfide.
0361-0128/98/000/000-00 $6.00 701

702 IOANNOU ET AL.
the data were not pressure corrected and thus represent min- Work by Delaney et al. (1987), Goldfarb and Delaney
imum trapping temperatures. The Th and salinity values indi- (1988), Fox (1990), Cathles (1993), and Butterfield and Mas-
cate a density of 0.73 ± 0.04 g/cm3 (1σ, n = 24) for hy- soth (1994) suggests that phase separation followed by the
drothermal fluids associated with the Olga Lake pluton. isolation of brine from vapor will occur where there is a series
of constricted and branching channels and laminar flow, ulti-
Discussion mately leading to the venting of vapor and gravitational accu-
mulation of brine deep within the hydrothermal system. This
Brine generation and segregation is the scenario suggested to explain the presence of saline-
Salinities that vary from that of normal seawater appear to rich and absence of vapor-rich fluid inclusions in quartz-
be the rule rather than the exception for modern submarine epidote veins of the Bell River Complex at Matagami. The
hydrothermal vents. Furthermore, salinity enrichments are subhorizontal orientation of the quartz-epidote–related hy-
more common than salinity depletions (Von Damm and drothermal cracking zone fractures (Ioannou and Spooner,
Bischoff, 1987; Bischoff and Rosenbauer, 1989). 2007) would have provided a good setting for effective gravi-
Under certain (low) water to rock ratios and P-T conditions, tational separation of phase-separated brine and vapor.
significant enrichments in the salinity of hydrothermal fluids Nested hydrothermal convection cells: An oxygen isotope
can be obtained through hydration processes (Hutchinson et profile of the Oman ophiolitic sequence indicates that two
al., 1980; Cathles, 1983; Macdonald and Fyfe, 1983). How- distinct, vertically separated hydrothermal systems operated
ever, to double the salinity of a fluid, hydration must consume during the formation of the fossil oceanic crust (Gregory and
50 percent of the initial seawater (Delaney et al., 1987), sug- Taylor, 1981). Nested cells are well known in fluid mechanics,
gesting that a five-fold increase over seawater salinity, as cal- a consequence of double diffusive convection (Griffiths,
culated for the primary fluid inclusions in VMS ore at 1981), which has many geologic applications, including the
Matagami, or the ten-fold increase calculated for primary flu- differentiation of magmas and mantle flow (Huppert and
ids in the hydrothermal cracking zone of the Bell River Com- Sparks, 1984), the layering of oceanic waters (Stommel et al.,
plex, is not feasible by this mechanism. 1956), and the behavior of black smoker plumes ejected into
Precipitation and dissolution of a Cl-bearing mineral phase the open ocean (Huppert and Turner, 1981).
has been suggested as a mechanism for variable vent fluid In the case of a two-cell VMS hydrothermal system, brine
salinity (Edmond et al., 1979). Experiments by Seyfried et al. in the lower cell undergoes recycling, whereas seawater in the
(1986) provided some support for this interpretation at tem- upper cell circulates in an approximately single-pass cycle
peratures near 425°C. However, the small variations in Cl with a relatively short residence time of likely less than 3 years
concentrations at lower temperatures raises questions about (Kadko and Moore, 1988; Rosenberg et al., 1993). The lower,
the effectiveness of this process at temperatures (<350°C) brine-dominated cell, characteristically dominated by fluids
more typical of VMS-forming systems. The Na/Cl ratios for with a higher density, temperature, and boiling point, pro-
vent fluids with elevated chlorinity are sharply discordant to vides a stable medium to convect heat from the thermal
the seawater evaporation trend and fluids of depleted chlo- boundary layer that occurs between convecting magma below
rinity (Bischoff and Rosenbauer, 1989), suggesting that the and convecting hydrothermal fluid above (Bischoff and
observed salinity variations are not produced solely by simple Rosenbauer, 1989; Cathles, 1993). The thermal boundary
reversible processes, as would be expected from precipita- layer is itself bounded by the hydrothermal cracking zone
tion-dissolution models. (Lister, 1974; Cathles, 1993; Nehlig, 1993; Ioannou and
Phase separation: Mixing altered seawater with a gas-de- Spooner, 2007); the zone in which hot plastic rock is
pleted brine has been proposed to explain the high salinities quenched by cooler convecting hydrothermal fluids. This
encountered in a number of modern submarine hydrother- quenching causes the plastic rocks to shrink and crack, pro-
mal systems (e.g., Von Damm and Bischoff, 1987; Cowan and ducing a highly permeable horizon for hydrothermal fluids to
Cann, 1988; Von Damm, 1988; Vanko et al., 1992). Phase sep- penetrate, obtain heat and metals, and undergo phase sepa-
aration deep within a hydrothermal system (600°–>700°C, ration (Lister, 1975; Fournier, 1987; Ioannou and Spooner,
60–120 MPa: Delaney et al., 1987; Kelley and Delaney, 1987; 2007). At Matagami, the near-lithostatic pressure gradient
Vanko, 1988) is the likely mechanism for producing the saline (see below) within the hydrothermal cracking zone of the Bell
brine (Delaney and Cousens, 1982; Bischoff and Pitzer, 1985; River Complex suggests that this region of the hydrothermal
Fournier, 1987; Butterfield et al., 1990). system was isolated from circulation above, and mass balance
At shallower depths and lower temperatures (<29.9 MPa models (Ioannou, 2004) suggest that, in addition to convec-
and <407°C: Bischoff and Rosenbauer, 1988) phase separation within the hydrothermal cracking zone, convection and
tion will generate a low-salinity vapor phase (bubbles) within subsequent metal leaching also occurred within the Watson
a continuous liquid phase of slightly increased salinity. How- Lake Group beneath the South Limb VMS deposits, support-
ever, the temperatures involved in the subcritical portion of ing the presence of a stacked two-cell system.
the H2O-NaCl system are insufficient to separate more than Field relationships indicate that the hydrothermal cracking
30 to 40 percent vapor under optimal adiabatic conditions zone of the Bell River Complex was separated from perme-
(Delaney, 1982; Hedenquist, 1984), resulting in a maximum able volcanic rocks of the Watson Lake Group by approxi-
residual liquid salinity less than 2 times that of seawater—well mately 650 m of less permeable upper Bell River Complex
below the ±30 wt percent NaCl equiv determined in this and gabbro, anorthosite, and granophyre (Ioannou and Spooner,
other studies (e.g., Kelley and Delaney, 1987; Vanko, 1988; 2007). Effective diffusion of heat and dissolved components
Kelley and Robinson, 1990). across such a wide low-porosity interface is questionable.
0361-0128/98/000/000-00 $6.00 702

However, given the high salinity (16.2 ± 4.7 wt %) encoun- of the two-phase mixture is greater than the one-phase fluid
tered in ore-hosted primary fluid inclusions at Matagami, one (Goldfarb and Delaney, 1988). This volume and buoyancy in-
genetic model compatible with our data is a leaky two-cell crease could hinder the migration of hydrothermal fluid into
convection system. In this model, a significant amount of the two-phase field (Kelley et al., 1992). Therefore, the high-
brine, produced by phase separation deep in the lower cell of est salinity that would likely be generated from a seawater
the hydrothermal system, was incorporated into the upper source would contain ~30 wt percent NaCl (Kelley et al.,
cell and mixed with heated seawater during convection. This 1992), nearly 15 wt percent NaCl lower than observed in
may have been episodic, in association with tectonic events, in some primary fluid inclusions from the hydrothermal crack-
order to maintain a near-lithostatic pressure gradient within ing zone of the Bell River Complex. This suggests that the flu-
the hydrothermal cracking zone. For a model Archean sea- ids underwent phase separation at pressures greater than 55
water salinity comparable to that of modern seawater (3.2 wt MPa (i.e., >hydrostatic) given that the salinity of a phase-sep-
% NaCl; Bischoff and Rosenbauer, 1985), the relative amount arated brine increases with the pressure under which the
of leakage can be estimated. For example, 0.6 kg of brine phase separation occurs.
(38.23 ± 1.85 wt % NaCl equiv) generated in the lower cell At Matagami, a lithostatic pressure gradient to a depth of 2
would be required to leak and mix with each kilogram of sea- km beneath the paleosea floor produces a maximum Pt of 90
water in the upper cell in order to produce a fluid with ~16.2 MPa in the hydrothermal cracking zone of the Bell River
wt percent NaCl. Applying the same reasoning to the mini- Complex, as defined above. At 90 MPa, seawater will undergo
mum (1.76 wt %) and maximum (23.30 wt %) microthermo- phase separation near 650°C (Figs. 7, 8), initially producing a
metrically determined salinities for ore-hosted fluids, we sug- brine (we note that phase separation above 600°C is well doc-
gest that the relative leakage varied temporally and spatially umented in some modern sea-floor hydrothermal systems:
between 0.0 and 1.3 kg of brine per kg of model seawater. In- e.g., Stakes and Vanko, 1986; Kelley and Delaney, 1987). A
clusions containing fluids with salinities less than 3.2 wt per- phase-separated brine produced under these temperature
cent document the mixing of seawater with vapor-rich phase and pressure conditions will contain approximately 47 wt per-
separated fluids from deeper within the system. Evidence for cent NaCl (Sourirajan and Kennedy, 1962)—higher than the
an Archean seawater significantly more saline than 3.2 wt per- mean salinity of the primary fluid inclusions in quartz-epidote
cent NaCl (see below) implies less leakage than that calcu- veins in the Bell River Complex but more interestingly, close
lated above. to the maximum salinity (44.1 wt % NaCl equiv) for the in-
clusion population. Minor mixing of the brine with seawater
P-T-X-ρ pathways of ore-related fluid that had not undergone phase separation is a reasonable sce-
Seawater origin: Although the exact mechanism by which nario in the hydrothermal cracking zone (e.g., Delaney et al.,
hydrothermal convection in the lower cell is initiated and 1987; Cowan and Cann, 1988; Kelley et al., 1992), ultimately
maintained is not fully understood, a likely scenario involves diluting the brine to, in this case, ~38 wt percent NaCl. Aside
the introduction of seawater deep into the crust during from mixing considerations, the general agreement between
episodic magmatic and/or tectonic events, where it can inter- measured and experimental salinity data independently sup-
act with magma, undergo phase separation, and produce a ports our estimate of a lithostatic pressure of 90 MPa for the
saline brine. hydrothermal cracking zone and an Archean water depth of
The maximum depth of hydrothermal fluid penetration 3,500 m at Matagami as originally suggested by MacGeehan
within the Bell River Complex is well constrained by geologic (1979). In addition, the evidence for a near-lithostatic pres-
reconstruction to have been 2 km beneath the paleosea floor sure regime within the hydrothermal cracking zone of the
(Ioannou and Spooner, 2007). A key parameter in modeling Bell River Complex supports the model of an independent
possible P-T-X-ρ pathways is recognition of whether pore and/or isolated (for the most part) convection cell at depth.
fluid pressures approximated hydrostatic or lithostatic condi- The transition zone, or rigidus as defined by Lister (1986),
tions (Kelley and Delaney, 1987). Hydrostatic conditions separating hydrostatic and lithostatic pressure regimes is
imply brittle deformation and a fracture network connected characterized by a steeper pore fluid pressure gradient than
to the sea floor. At Matagami, a hydrostatic pressure gradient would characterize either a purely hydrostatic or purely litho-
to a depth of 2 km beneath the paleosea floor produces a static pressure gradient (Kelley and Delaney, 1987). This
pressure of 55 MPa within the hydrothermal cracking zone of rigidus produces a significant offset in the two-phase curve.
the Bell River Complex (given a model water depth of 3,500 In any active magma-hydrothermal system, the position of the
m and ρ = 1.0 g/cm3; MacGeehan, 1979). At 55 MPa, a 3.2 wt rigidus is not stable. A downward shift in the rigidus results in
percent NaCl solution will undergo phase separation at substantial decompression of fluids that had been at lithosta-
500°C (note the critical point for seawater occurs at 407°C, tic pressure. As a result, one-phase fluids could decompress,
29.9 MPa: Bischoff and Rosenbauer, 1988), producing a ~25 inducing two-phase separation without significant vertical mi-
wt percent NaCl brine and a highly expanded vapor phase gration or temperature change (Goldfarb and Delaney, 1988).
with a salinity of ~2.3 wt percent NaCl (Sourirajan and Such decompression is precisely what is expected with the
Kennedy, 1962; Fig 7). The mass ratio of these phases can be onset of hydrothermal cracking as ductile rocks become brit-
estimated using the lever rule at approximately 4 percent tle (Lister, 1974, 1986; Kelley and Delaney, 1987), making the
brine and 96 percent vapor. Because NaCl has a large nega- hydrothermal cracking zone of the Bell River Complex an
tive partial molar volume (Copeland et al., 1953; Urusova, ideal location for phase separation to have taken place.
1975) and the low-salinity vapor phase is significantly less Note that seawater will undergo phase separation to pro-
dense than the homogenous parental fluid, the total volume duce a 38 wt percent NaCl brine at approximately 75 MPa
0361-0128/98/000/000-00 $6.00 703

704 IOANNOU ET AL.
o
Temperature ( C) Salinity (wt.% NaCl)
0 100 200 300 400 500 600 700 800 0 10 20 30 40 50
0
3.2 wt.% NaCl
5.3 wt.% NaCl

o
300 C
L+
L+
o
350 C
V(
H
3 .2
o
w t.
20 400 C
%N
L+
V
aCl)
+H
o
C
450
A E A E A E
0.0
40
Sup
erc r
o
C WLG
500
it
S1 D S1 S1
ic al
D
Pressure (M Pa)
Pha
Depth Below Seafloor (km )

se
o C
Sep
0
60 55
1.0
arati
Granite Solidus
on
Cur
BRC
v e (3
o C
.2 w
0
60
80
t.%
NaC
S2 C S2
C
l)
M,M' S2 HCZ C
2.0
wt.% NaCl)
B,B' B,B'
L+H (38.23 wt.% NaCl)
MB o
C B' M'
620 BRC
100
onvec tion
Tm ice = T h (38.23
cC
C
o
C
ti
650
ma
0
70
M ag
0 C
o
67
120 3.0
FIG. 7. Schematic model of the P-T-X pathway for VMS-related hydrothermal fluids at Matagami. The model is con-
structed using the H2O-NaCl system data of Sourirajan and Kennedy (1962), Bischoff (1991), Bodnar (1992), and Bodnar
(1994). Black points indicate conditions coincident with the phase separation surface; white points indicate projected posi-
tions. Black stars mark the critical point of seawater (3.2 wt % NaCl: Bischoff and Rosenbauer, 1988). The bracketed lines
on either side of points (C) and (E) mark the range and standard deviation (1σ) for the fluid inclusion population trapped
under the respective conditions. Seawater brine model: (A) Seawater (3.2 wt % NaCl) was introduced deep into the crust.
At 90 MPa (2.0 km beneath the sea floor) the seawater reached 650°C (B), subsequently undergoing phase separation which
produced a brine (B') containing ~47 wt percent NaCl. The high density of the brine restricted it to convection within the
hydrothermal cracking zone. Phase separation and mixing with lower temperature, nonphase separated seawater (S2) cooled
and diluted the brine to ~300° to 500°C and ~38 wt percent NaCl (C). It is under these conditions that many of the primary
fluid inclusions in quartz-epidote veins of the Bell River Complex (BRC) trapped single-phase (liquid) fluids. Some of the
lower temperature fluid from this population was trapped close to the halite liquidus, resulting in the observed Tm(salt) > Th(L-
V(L)). Leakage of the brine into the upper cell resulted in further mixing with seawater (S1). The mixing probably occurred
within the lower half of the Watson Lake Group (WLG) (D). The resultant buoyant one-phase (liquid) hydrothermal fluid
subsequently rose toward the sea floor, where some of it was trapped as primary fluid inclusions in VMS ore at temperatures
of 160° to 290°C and a pressure of 35 MPa (E). Note the dashed 5.3 wt percent NaCl line marked in the pressure vs. salin-
ity plot. At 75 MPa a fluid of this composition will undergo phase separation producing a ~38 wt percent NaCl brine—with-
out subsequent mixing (B'-C) with nonphase separated seawater as above. Magmatic brine model: At 90 MPa a magmatic
fluid exsolved from a granitic melt at approximately 670°C (granite solidus after Kelley and Robinson, 1990) would be within
the two-phase field of the H2O-NaCl system (M). Phase separation of the magmatic fluid produces a brine near 50 wt per-
cent NaCl (M'). The subsequent P-T-X-ρ history of the fluid would have then followed that of the seawater-derived fluid de-
scribed above, beginning with the mixing of cooler, nonphase separated seawater (S2 and C). Refer to text for discussion.
HCZ = hydrothermal cracking zone
and 575°C, suggesting that the transition into the hydrother- (90 MPa) and mixing with lower temperature seawater, low-
mal cracking zone may have been characterized by a mixed ers the temperature of the brine, bringing it back to the liq-
and/or varying hydrostatic and/or lithostatic pressure regime. uid field. Such mixing is suggested by the positive slope of the
Such a scenario alleviates the need for subsequent dilution to best fit through the primary fluid inclusion data from the
produce the average salinity of the primary fluid inclusions in quartz-epidote veins in salinity versus Tt space (Fig. 6). It is
quartz-epidote veins of the Bell River Complex. under these conditions (~300°–500°C, 90 MPa) that many of
Regardless of the exact P-T conditions at which phase sep- the quartz-epidote vein-hosted inclusions likely trapped one-
aration takes place, the low density of the phase-separated phase (liquid) fluids (Figs. 7, 8). Some of the lower tempera-
vapor-rich phase causes it to rise rapidly through the hydro- ture fluids of this population trapped close to the halite liq-
thermal system, ultimately mixing with seawater and/or dis- uidus resulted in the observed Tm(salt) > Th(L-V(L)) (Fig. 7).
charging at the sea floor. In contrast, the high density of the Vapor-rich inclusions associated with halite-bearing (i.e.,
brine-rich phase (Fig. 8) restricts it to convection within the brine) inclusions in which Tm(salt) > Th(L-V(L)) cannot represent
hydrothermal cracking zone. Convection at constant pressure an immiscible pair because of phase equilibrium constraints
0361-0128/98/000/000-00 $6.00 704

1.8
50 MPa
1.6
1.4
r ve 70
n Cu
1.2 turatio
16 eq.w t.%
Halite Sa 60
NaCl-CaCl
A ρLC
Density (g/c m )
2
S1
3
1.06
ρORE 1.03 E* 50
1.0 1.00 C 38 e
q.wt.
(3 5 M % 50
l (90 40
pure H2 O NaC
E Pa)
D* M pa
ρUC 0.90 30 )
M'
0.88 r ve
D (5 0 M Pa) 20 Cu 40 B'
0.8 S2
io n
e
ur v )
t
ara
10
wt. n C t ic
p
% tio em a
Phas e Se
Na a
Cl ar sc h
( 9 0 M Sep
0.6
;
Pa
Phas e
0.4
0.2
B M
0.0
0 100 200 300 400 500 600 700 800
o
Temperatur e ( C)
FIG. 8. Schematic ρ-T pathway model for VMS-related hydrothermal fluids at Matagami. The model is constructed using
data for the H2O-NaCl system data at 50 MPa from Khaibullin and Borisov (1966), Urusova (1975), Potter and Brown (1977),
Hilbert (1979), and Tanger and Pitzer (1989). Black points indicate conditions in the plane of the figure (50 MPa), white
points indicate projected positions. LC = lower cell, UC = upper cell. The bracketed lines on either side of points (C) and
(E*) mark the range and standard deviation (1σ) for the fluid inclusion population trapped under the respective conditions.
Seawater brine model: (A) Seawater (3.2 wt % NaCl; model salinity) was introduced deep into the crust. At 90 MPa the sea-
water reached 650°C (B), subsequently undergoing phase separation which produced a brine (B') containing ~47 wt percent
NaCl. The high density of the brine restricted it to convection within the hydrothermal cracking zone. Phase separation and
mixing with lower temperature, nonphase-separated seawater (S2) cooled and diluted the brine to ~300° to 500°C and ~38
wt percent NaCl (C). It is under these conditions (ρLC = 1.06 g/cm3) that many of the primary inclusions in quartz-epidote
veins in the Bell River Complex trapped single-phase (liquid) fluids. Leakage of the brine into the upper cell resulted in its
mixing with seawater (S1). The mixing probably occurred within the lower half of the Watson Lake Group (~50 MPa) near
350°C (D). The resultant buoyant (ρUC = 0.88 g/cm3) one-phase (liquid) hydrothermal fluid subsequently rose toward the
sea floor, where some of it was trapped as primary fluid inclusions in VMS ore at conditions of 160° to 290°C and 35 MPa
(ρORE = 1.00 g/cm3, E). Note that points E and D are modeled for a fluid containing only NaCl and, therefore, do not rep-
resent the Matagami ore fluids accurately. Points D* (ρUC = 0.90 g/cm3) and E* (ρORE = 1.03 g/cm3) represent the same con-
ditions but in the system H2O-NaCl-CaCl2 (based on the data of Zhang and Frantz, 1987). Note the density increase
(0.02–0.03 g/cm3) caused by the inclusion of CaCl2. Magmatic brine model: (M) At 90 MPa a magmatic fluid exsolved from
a granitic melt at approximately 670°C (granite solidus after Kelley and Robinson, 1990), placing it within the two-phase field
of the H2O-NaCl system. Phase separation of the magmatic fluid produced a brine near 50 wt percent NaCl (M'). The sub-
sequent P-T-X-ρ history of the fluid would have then followed that of the seawater brine origin fluid described above be-
ginning with the mixing with cooler, nonphase separated seawater (S2 and C). Refer to text for discussion.
(Cloke and Kesler, 1979; Roedder, 1984). However, in the normal offset quartz-epidote veins orthogonal to layering in
case of the primary inclusions in quartz-epidote veins, the the Bell River Complex (Ioannou and Spooner, 2007), re-
brine resulting from phase separation underwent further sulted in their mixing with seawater. This mixing likely oc-
modification (a temperature decrease back to the liquid field) curred within the lower Watson Lake Group (≤50 MPa) at
prior to trapping. A similar process for the trapping of brines temperatures of approximately 350°C (Figs. 7, 8), consistent
in deep-seated plutonic rocks of the Oceanographer Trans- with mass-balance calculations (Ioannou, 2004) and footwall
form, Mid Altlantic Ridge, has been suggested by Vanko et al. δ18O and fluid inclusion data of Costa et al. (1983) from the
(1992). Their strikingly similar fluid inclusion temperature Mattagami Lake mine. The less permeable nature of the up-
(±340°C) and salinity (40 wt % NaCl-CaCl2 equiv) to the flu- permost part of the Bell River Complex likely did not allow
ids from the hydrothermal cracking zone of the Bell River pervasive seawater penetration into it (Ioannou and Spooner,
Complex support the proposed phase separation at depth be- 2007). The resultant buoyant hydrothermal fluid subse-
neath the sea floor. quently rose toward the sea floor, where some of it was
Leakage via turbulent entrainment and/or episodic tectonic trapped as primary fluid inclusions in VMS ore under condi-
release of the lower cell brines into the upper cell, along tions of approximately 160° to 290°C and 35 MPa (Figs. 7, 8).
0361-0128/98/000/000-00 $6.00 705

706 IOANNOU ET AL.
Initial inspection of the fluid inclusion data indicates that The primary inclusions in the quartz-epidote veins from the
the fluids in the lower cell were of similar density (1.06 ± 0.03 hydrothermal cracking zone of the Bell River Complex con-
g/cm3, C in Fig. 8) to the ore-hosted fluids in the upper cell tain no sylvite daughter crystals at room temperature, sug-
(1.03 ± 0.05 g/cm3, E and E* in Fig. 8). However, conditions gesting that the fluids may not be magmatic in origin. Sterner
deeper within the Watson Lake Group, likely ±350°C, ±50 et al. (1988) noted that the absence of sylvite can be attrib-
MPa (D in Fig. 7 and D and D* in Fig. 8), and a correspond- uted to its metastable behavior at room temperature. How-
ing fluid density of ~0.90 g/cm3 (D and D* in Fig. 8), would ever, supercooling the inclusions below –130°C did not nu-
have been consistent with a two-cell convection model. cleate visible sylvite daughters.
With the eventual cooling of the magmatic heat source and Theoretically it is likely that the deep-seated hydrothermal
subsequent waning of hydrothermal activity, lower tempera- brines in the hydrothermal cracking zone of the Bell River
ture fluids, dominantly of seawater origin, circulated into the Complex are the product of both magmatic and seawater-de-
hydrothermal system. Modified by water-rock interaction rived fluids (e.g., Wolfgang et al., 2001). For example, a mag-
(e.g., Hardie, 1983), yet not phase separated, these fluids matic fluid contribution has been proposed for elevated fluid
sealed off the remnant permeability of the fracture network salinities of the Kuroko deposits by a number of authors (e.g.,
in the hydrothermal cracking zone and were locally trapped Sato, 1977; Urabe and Sato, 1978; Urabe et al., 1983). Direct
as secondary fluid inclusions (Th = 215°–271°C, ~9 wt % magmatic contributions, up to 25 percent, to the hydrother-
NaCl equiv) hosted within the quartz-epidote vein material of mal fluid have been estimated for various VMS systems (e.g.,
the Bell River Complex, analogous to the proposed origin of Sawkins and Kowalik, 1981; Urabe, 1987; Cathles, 1993; Yang
type I fluid inclusions in gabbros associated with hydrother- and Scott, 1996). Isotopic data support this interpretation (de
mal activity near the intersection of the Kane fracture zone Ronde, 1995). For example, Costa et al. (1983) use stable iso-
and Mid-Atlantic Ridge (Kelley and Delaney, 1987) and in tope data to suggest a magmatic contribution of up to 15 per-
the upper level plutonic sequence of the Troodos ophiolite, cent for the Mattagami Lake mine. The authors note that
Cyprus (Kelley and Robinson, 1990; Kelley et al., 1992). such a magmatic contribution could have supplied significant
Magmatic origin: An additional possible origin for brine amounts of metals to the hydrothermal system.
components in the deepest levels of a hydrothermal system, At Matagami, the granophyric carapace of the Bell River
such as the hydrothermal cracking zone of the Bell River Complex shows little sign of brittle failure, although the unit
Complex, is as an exsolved magmatic component. This com- is poorly exposed in outcrop. Furthermore, a lack of signifi-
ponent may be exsolved directly as a brine from late-stage cant visible CO2 in the hydrothermal system, in the form of
melts in low-pressure magmatic systems (e.g., Campbell et CO2-rich fluid inclusions such as located in the synvolcanic
al., 1981; Kelley et al., 1992; Shinohara, 1994) or, at higher 2700.8 +2.6/–1.0 Ma Flavrian pluton in the Noranda district
pressures, as a fluid that subsequently undergoes phase sepa- (Goldie, 1979; Mortensen, 1993b; L. Weiershäuser pers.
ration near the brittle-ductile transition zone to produce a commun., 2003) suggests that magmatic contribution, if any,
brine (e.g., Kelley and Delaney, 1987; Kelley and Robinson, was minor. However, magmatic differentiation leading to the
1990; Vanko et al., 1992; Kelley, 1997). Subsequent segrega- formation of the granophyric carapace of the Bell River Com-
tion of a phase-separated brine at depth, as the vapor phase plex may, in part, explain this lack of significant CO2, given
migrates to shallower crustal levels (Goldfarb and Delaney, the low solubility of CO2 in felsic differentiates (Gerlach,
1988), could produce the high-salinity inclusions found in the 1989).
hydrothermal cracking zone of the Bell River Complex at In general, the large volumes of seawater associated with
Matagami and in the deepest portions of other VMS hy- VMS hydrothermal systems readily obscure any magmatic
drothermal systems (e.g., Kelley and Delaney, 1987). signatures. The permeable parts of the system include 1,000
At 90 MPa, a magmatic fluid will exsolve from a granitic m of Watson Lake Group volcanic rocks (ρ = 2.600 t/m3) and
melt at approximately 670°C (Kelley and Robinson, 1990), 350 m thickness of the hydrothermal cracking zone in the Bell
placing it within the two-phase field of the H2O-NaCl system. River Complex (ρ = 2.900 t/m3). Based on the weighted aver-
At Matagami, phase separation of such a magmatic fluid age density of the rocks, the system encompasses 3,615 Mt of
would produce a brine near 50 wt percent NaCl (Fig. 7). The permeable rock per km2 (1,350 × 1,000 × 1,000 × 2.678
subsequent P-T-X-ρ history of the fluid would then follow t/m3). Assuming a fluid/rock mass ratio of 0.4 over the entire
that of fluids of seawater origin described above, beginning permeable parts of the Matagami hydrothermal system (Wat-
with the mixing with cooler, nonphase-separated seawater son Lake Group and hydrothermal cracking zone of the Bell
(Figs. 7, 8). A model involving a magmatic fluid as the source River Complex), each square kilometer of the hydrothermal
of brine found in the hydrothermal cracking zone of the Bell system contained 1,446 Mt of fluid (Norton and Knight,
River Complex would account for the current lack of an iden- 1977). A maximum contribution from magmatic fluids of 15
tified recharge pathway(s) for Archean seawater that might percent estimated by Costa et al. (1983) would correspond to
have migrated into the hydrothermal cracking zone (Ioannou a magmatic fluid input of 217 Mt/km2. A 1 percent magmatic
and Spooner, 2007). However, we emphasize the relatively contribution would correspond to a fluid input of 15 Mt/km2.
poor outcrop exposure at Matagami. More detailed work, such as fluid inclusion bulk analytical
The presence of sylvite daughter crystals in pluton-hosted gas- and ion-chromatography (GC/IC) work (cf. Bray et al.,
brine inclusions has been linked to a magmatic fluid origin 1991; Channer et al., 1999) is needed to properly assess the
(Kelley and Delaney, 1987; Kelley et al., 1992), as magmatic role of a magmatic contribution and quantify the concentra-
aqueous fluids in equilibrium with a fractionated silicate melt tion of CO2 and other volatile species in the hydrothermal
are characteristically enriched in potassium (Burnham, 1997). fluid.
0361-0128/98/000/000-00 $6.00 706

Metal solubility The solubility of Cu as chloride complexes is more tempera-

The mean trapping temperature of Matagami ore fluid es- ture dependent than that of Zn (e.g., Crerar and Barnes, 1976;
timated here (Tt = 208° ± 32°C) falls within the lower limits Finlow-Bates, 1980; Franklin et al., 1981; Large, 1992; Seyfried
of the temperature range (~200°–400°C) cited for modern et al., 1999; Shosa, 2000), and temperature likely controlled the
and ancient VMS systems (Scott, 1997), which raises ques- solubility of Cu in the Matagami ore fluid, despite the high
tions about the ability of the fluid to transport Zn and Cu. salinity. For example, the Cu solubility data of Crerar and
Dissolution and transport of Zn and Cu in VMS systems are Barnes (1976) indicate that the Matagami fluids were capable of
dominated by chloride complexes (Ruaya and Seward, 1986; dissolving ~3 ppm Cu (Fig. 9B), less than half the solubility in
Bourcier and Barnes, 1987; Seward and Barnes, 1997). De- higher temperature (~350°C), lower salinity (~4.0 wt % NaCl
spite the low mean temperature, the high salinity (16.2 ± 4.7 equiv) fluids associated with many VMS systems (Scott, 1997).
wt %) of the ore fluids would have made them very effective Conditions for Cu (and Zn) chloride complexing would
at transporting significant amounts of Zn. Hanor (1996) have been more favorable in the hydrothermal cracking zone
showed that several properties of lower temperature of the Bell River Complex (Cu solubility >45 ppm at
(≤150°C) basinal fluids (including low pH, predominance of (~370°C, ~38 wt % NaCl: Crerar and Barnes, 1976; Camp-
ZnCl42– complexes, and the strongly nonideal behavior of chlo- bell et al., 1981). However, subsequent mixing of the deep
ride) combine to increase the solubility of sphalerite in a non- brine with seawater, well below the VMS ore deposits, would
linear manner by many orders of magnitude as salinity in- have caused significant precipitation of most dissolved Cu be-
creases at fixed temperature and aH2S. Applying the ore-fluid fore the fluid reached the sea floor.
data of this study to the Zn solubility data of Bourcier and Zn and Cu have been shown to form stable bisulfide com-
Barnes (1987) indicates that the Matagami fluids were capa- plexes in laboratory studies (Barnes, 1965; Crerar and Barnes,
ble of dissolving ~5 × 10–4 m (30 ppm) Zn (Fig. 9A), nearly 1976). However, there is still considerable uncertainty re-
identical, for example, to the solubility of Zn in the higher garding the stoichiometries and thermodynamics of these
temperature (~260°C), lower salinity (~4.0 wt % NaCl equiv) complexes (Wood and Samson, 1998) which, to be of eco-
fluids of the Kuroko VMS deposits, Japan (Pisutha-Arnond nomic significance, require an H2S activity ~2+ orders of
and Ohmoto, 1983). This ~30-ppm estimate is also consistent magnitude higher than is commonly associated with VMS sys-
with zinc solubilities calculated from Yardley (2005) for a tems (10–2–10–3 m total reduced sulfur: Barnes, 1979). Bisul-
208°C fluid with a 2.7-m salinity (35 ppm Zn): log(Zn/Cl) = – fide complexing requires conditions that are more alkaline
(1.781/T) ±1, where concentration ratios are molar and T = and lower in temperature than those associated with most
temperature (Kelvin). modern (Scott, 1997) and ancient (e.g., Pisutha-Arnond and
With ore fluids reaching 285°C and 23.30 wt percent NaCl- Ohmoto, 1983) VMS systems.
CaCl2 equiv, the Zn-rich nature of the Matagami deposits is
not surprising. At 208°C, the chloride complexes ZnCl–3 and Fluid end-member compositions
ZnCl42– equally dominate the Zn solubility in the ore fluid As discussed above, fluid inclusion microthermometry indi-
(Bourcier and Barnes, 1987). cates that some of the Matagami fluids contain significant
10 1
-2
10
(A) (B)
5.0 m
4
10 -1 10
2
10
-3
10
Copper Solubility (ppm)

Copper Solubility ( m )
5. 0 m
Zinc Solubility (ppm)
2
Zinc Solubility ( m )
10 -3 10
m
0.5 olate
d)
(interp 1
2 .7 m 10
10 -5 0. 1m 10
0
10 -4 0.5 m
d)
ola te
erp
(int
2.7
m 0.1 m
10
-7 10 -2
10 0
10 -5
10
-9 10 -4
10 -11 10 -6
100 150 200 250 300 350 100 150 200 250 300 350
Temperatur e (oC) Temperature ( oC)
FIG. 9. (A). Zn solubility vs. temperature plot (modified from Bourcier and Barnes, 1987; pH = 5.0). The dashed lines
correspond to a 2.7 ± 0.9 m NaCl-CaCl2 range (1σ salinity variation from the 2.7 m mean salinity of primary fluid inclusions
in VMS ore), and a temperature range of 208° ± 32°C (1σ variation in the mean trapping temperature of primary fluid in-
clusions). Note the high salinity of the Matagami VMS ore fluid more than compensates for its low temperature in terms of
Zn transport. (B). Cu solubility vs. temperature plot (based on the data of Crerar and Barnes, 1976). Note that despite the
high salinity of the Matagami ore fluid, its relatively low temperature limits significant Cu transport.
0361-0128/98/000/000-00 $6.00 707

708 IOANNOU ET AL.
proportions of salt species other than NaCl; likely CaCl2. The undergoing phase separation at 90 MPa (620°C) will produce
presence of CaCl2 is well documented in VMS hydrothermal a brine containing ~38 wt percent NaCl, exactly what is seen
systems (e.g., Michard et al., 1984; Von Damm et al., 1985; in primary fluid inclusions in the quartz-epidote veins in the
Vanko, 1988; Vanko et al., 1992). Hardie (1983) has shown Bell River Complex (Fig. 7). Similar Cl-rich fluids have been
theoretically that seawater reacting with basalt at 275°C may found in ore-hosted primary fluid inclusions from the
become chemically modified to a CaCl2 brine by Na+-Ca2+ Archean (2.72 Ga) Kidd Creek VMS deposit, Ontario (5.7 ±
cation exchange, a finding supported by experimental seawa- 0.5 wt % NaCl equiv, 1σ, 4.7–6.3, n = 29: Schandl and
ter-basalt and seawater-diabase reactions (Mottl and Holland, Bleeker, 1999; 7.0 ± 1.5 wt % NaCl equiv, 1σ, 4.1–12.1, n =
1978; Seyfried and Bischoff, 1981; Seyfried and Mottl, 1982). 49: S.E. Ioannou and E.T.C. Spooner, unpub. data).
The presence of basalts in the footwall Watson Lake Group at
Matagami makes such reactions one plausible mechanism for Fluid-density considerations
generating CaCl2-rich fluids (Costa et al., 1983). Costa et al. (1983) interpreted their Mattagami Lake mine
Vanko (1988) reported a molar Na/Ca ratio of 3.6 to 8.4 fluid inclusion data and the presence of abundant talc in the
(2.9–7.3 by wt) for quartz-hosted fluid inclusions in metagab- ore and alteration pipe in terms of a sea-floor brine pool
bros from the Mathematician Ridge, East Pacific, and attrib- model analogous to the Red Sea. Zierenberg (1984) pointed
utes the Ca-rich nature of the fluids to seawater-basalt reac- out that given Costa et al.’s (1983) salinity and Th data, the hy-
tions such as the albitization of plagioclase at greenschist drothermal fluids would not pond on the sea floor as a brine
facies conditions (Seyfried, 1987; Berndt et al., 1988; Bowers pool but would rise as a buoyant hydrothermal plume into the
et al., 1988). Seawater has a molar Na/Ca ratio of 45.9 (26 by Archean seawater column. Costa et al.’s (1984) reply noted
wt: Bischoff and Seyfried, 1978). Ore fluids at Matagami are that further fluid inclusion studies found high-salinity, liquid-
characterized by a maximum NaCl/CaCl2 weight ratio of 1.22 dominant inclusions, including three-phase liquid-vapor-solid
(XNaCl = 0.55), which translates into a molar Na/Ca ratio of (halite) inclusions with Th(L-V(L)) ≈ Tm(salt) values of 240° to
2.3, lower than that found at the Mathematician Ridge 280°C, corresponding to a salinity of 35 to 38 wt percent
(Vanko, 1988) and in modern hydrothermal systems (e.g., NaCl equiv. However, a comprehensive breakdown of these
4.5–5.8 by wt: Edmond et al., 1979; Michard et al., 1984; Von data is not available.
Damm et al., 1985; Von Damm and Bischoff, 1987), yet sim- Our data confirm the presence of high-salinity ore fluids at
ilar to synvolcanic VMS-related fluids from the 070 faults of Matagami. A significant portion of our data show that the ore
the Archean Corbet mine, Noranda, Quebec (Farr, 1984). fluid was more dense than present-day seawater (~1.04 g/cm3
Fluid inclusion bulk analyses of 3230 Ma sea-floor hydrother- at 2°C for 3.2 wt % NaCl, regressed from the data of Potter
mal fluids in ironstone pods (de Ronde et al., 1997a) and in and Brown, 1977), as well as a higher salinity, hotter seawater
host rocks (Cloete, 1994, 1999) of the Barberton greenstone postulated for the Archean (~1.03 g/cm3 at 80°C for 5.3 wt %
belt, South Africa, indicate that Archean seawater was signif- NaCl, regressed from the data of Potter and Brown, 1977),
icantly richer in Ca than its present-day equivalent. These re- and could have formed a brine pool, consistent with the
sults are consistent with work by Walker (1983) and provide a strong stratigraphic control on sulfide mineralization at
possible explanation for the high Ca content of the Matagami Matagami, with most orebodies located at (within) the top of
(and Noranda) ore fluids. This finding is further supported by the Watson Lake Group, typically overlain by the Key Tuffite.
data from well-preserved 2.7 Ga fluid inclusions related to Sulfide quenching in a bottom-hugging brine pool provides a
Archean seawater-rock interaction from the Ben Nevis area of relatively efficient trapping mechanism for metals in the hy-
the Abitibi greenstone belt. These inclusions, hosted in inter- drothermal fluid (Solomon et al., 2002). A high Zn/Cu ratio,
stitial and vesicle quartz, contain fluids ranging from 4 to 22 characteristic of the Matagami district, is consistent with
wt percent CaCl2 equiv (Te(ice) = –45 to –30°C: Weiershäuser metal deposition via brine pool precipitation (Solomon et. al.,
and Spooner, 2005). 2004). However, our density data also indicates fluids less
Evidence for an Archean seawater composition different dense than ambient seawater, suggesting that buoyant hy-
from that of the modern oceans is not limited to Ca content drothermal plumes were also present (Fig. 10), consistent
but includes elevated concentrations of I, NH4, Sr, Na, and Br with variable fluid density suggested by Green et al. (1981) in
(de Ronde et al., 1997a, 2002). Channer et al. (1997) and other VMS systems.
Gutzmer et al. (2003) show that the Br-/Cl– and I–/Cl– ratios
of seawater were higher in the Archean, consistent with man- Hanging-wall quartz-epidote vein fluids
tle buffering. In addition, a number of workers propose that The wide range of salinity from 5.0 to 32.4 wt percent
seawater was significantly warmer, with temperatures ranging NaCl-CaCl2 equiv, high apparent CaCl2 (Te(min) = –51°C, max
from 30° to 80°C (Knauth and Lowe, 1978; Costa et al., 1980; XNaCl = 0.06) in quartz-hosted fluid inclusions, and relatively
Ohmoto and Felder, 1987; de Ronde et al, 1997b), and Hol- low Th (quartz = 124°C; epidote = 162°C) distinguishes the
land (1972) suggested that it was also more reducing; a sug- postore hanging-wall quartz-epidote vein-hosted fluids from
gestion which is still a topic of debate (e.g., Phillips et al., the ore-related fluids described above. The presence of pil-
2001). Calculated end-member compositions for Archean low structures throughout the Wabassee Group indicates that
seawater from the Barberton indicate a Cl concentration it was deposited subaqueously. Reaction of seawater at ele-
~165 percent that of the modern oceans (de Ronde et al., vated temperatures with at least 2 to 6 km (up to 13 km) of
1997a) that, in addition to phase separation, could partly Wabassee Group basalts (Mottl and Holland, 1978; Seyfried
explain the high salinities measured in a number of the and Bischoff, 1981; Hardie, 1983) could in part explain the
Matagami fluids. We note that a 5.3 wt percent NaCl fluid origin of the Ca-rich hanging-wall brines. Likewise, the brines
0361-0128/98/000/000-00 $6.00 708

1.6
35 MPa
3 )
3 1.2
Density (g/cm
Density (g/cm )
3 o
1.04 g/cm (3.2 wt.% NaCl @ 2 C)
3 o
1.03 g/cm (5.3 wt.% NaCl @ 8 0 C)
0.8
5.3
3. 2 wt.
w t. %
Na
%N Cl
aC
l
0.4
0 100 200 300 400 500
oo
Tt (( C)
C)
Tt
FIG. 10. Density vs. trapping temperature (Tt) plot for the primary fluid inclusion data from VMS ore; the bracketed lines
on either side of the mean VMS ore fluid density (white data point) indicate the standard deviation (1σ) for the data set. Con-
stant salinity curves are determined from a regression of the data of Potter and Brown (1977). A significant portion of our
data show that the ore fluid was more dense than present-day seawater (~1.04 g/cm3 at 3.2 wt % NaCl and 2°C), as well as
a higher salinity and/or temperature seawater postulated for the Archean (~1.03 g/cm3 at 5.3 wt % NaCl and 80°C), and could
have formed a brine pool. However, our data also indicate the presence of fluids less dense than ambient seawater, suggest-
ing that buoyant hydrothermal plumes were also present.
may reflect a more saline Archean seawater. Similar temper- would have had to traverse and react with at least 2 km of
ature (Th = 150°–160°C) and salinity (0.7–26 wt % CaCl2 Wabassee basalts prior to interaction with the Watson Lake
equiv) data from quartz-epidote veins of the Ben Nevis area Group. Seawater not circulated deeply, thus remaining within
of the Abitibi greenstone belt were interpreted as an area of the Wabassee Group, would not react with as much basalt
distal and/or passive seawater-rock interaction (Weiershäuser (i.e., lower effective water/rock ratio), becoming only moder-
and Spooner, 2005). These data support a high and variable ately enriched in Ca (<20.4 wt % NaCl-CaCl2 equiv). Given
Archean seawater salinity and a direct association between the relatively high density (average 1.12 g/cm3 in quartz
the hanging-wall quartz-epidote veins at Matagami and VMS- hosted inclusions) of these Ca-rich fluids, it is unlikely that
related processes. A scenario of postore or waning stage of hy- they reached the sea floor.
drothermal activity is supported by (1) the presence of multi-
ple exhalative tuffite layers throughout the Wabassee Group, Fluids related to the Olga Lake pluton
(2) bulk-rock geochemistry, XRD, and δ18O alteration signa- The quartz-vein hosted fluids associated with the Olga Lake
tures extending from the VMS ore deposits into the hanging granitoid pluton plot separately from other fluids in tempera-
wall (Carr, 2004; Ioannou, 2004), and (3) the presence of epi- ture-salinity space (Fig. 6), not surprisingly as the intrusion of
dote patches and veins in the hanging wall interpreted to rep- the pluton postdates the VMS by ~32 m.y. Presumably the
resent hydrothermal fluid channels (Harrigan and MacLean, higher temperature (>Th of 340° ± 20°C, depending on pres-
1976). In addition, Vervoort et al. (1993) interpret relatively sure) reflects the thermal influence of the pluton. Microther-
young (2647 ± 14 MaPb-Pb) model ages based on whole-rock mometrically, there is no indication of significant concentrations
Pb isotope data for the Wabassee Group as indicating the per- of Ca in the fluid (Te(min) = –14°C). At 2.5 times the salinity of
vasive circulation of hydrothermal fluids, derived from the contemporary seawater, these fluids may represent either the
dewatering of the lower parts of the newly formed thickened product of phase separation elsewhere in the system or a par-
Archean crust, 50 to 150 m.y. after this crust was assembled ticular sample of higher salinity modified Archean seawater.
and the related VMS deposits were formed.
Secondary fluid inclusions in massive sulfide-related quartz Conclusions
have temperatures and salinities similar to the fluids in hang- Microthermometric measurements on ore-hosted primary
ing-wall quartz-epidote veins, suggesting a similar origin. The liquid-vapor inclusions from four south-limb deposits at
lack of ore-hosted secondary inclusions with salinities less Matagami indicate that the VMS ore fluid was highly saline
than 20.4 wt percent NaCl-CaCl2 equiv could indicate that (16.2 ± 4.7 wt % NaCl-CaCl2 equiv, 1σ, n = 230) and of mod-
seawater-basalt reaction was occurring in areas of both erate temperature (Tt = 208° ± 32°C, 1σ, n = 230). A fluid
recharge and discharge. In the case of the former, seawater with these characteristics is able to transport ~5 × 10–4 m (30
0361-0128/98/000/000-00 $6.00 709

710 IOANNOU ET AL.
ppm) Zn as ZnCl–3 and ZnCl42–. However, the low temperature presence of pillow structures throughout the Wabassee
of this fluid precludes efficient Cu transport (only ~3 ppm), Group indicates that these rocks were deposited subaque-
which may in part explain the relatively Cu poor nature of the ously and therefore seawater would have continued to circu-
Matagami deposits. Calculated densities of this ore fluid, as late through the hanging-wall rocks after deposition of the
high as 1.10 g/cm3, are consistent with a bottom-hugging massive sulfide deposits. Reaction, at elevated temperatures,
brine model. However, a subset of the data indicates a fluid between seawater and the Wabassee Group basalts could ex-
less dense than ambient seawater, suggesting that buoyant hy- plain the origin of the Ca-rich hanging-wall brines. Likewise,
drothermal plumes were also present, consistent with geo- Archean seawater significantly more saline than modern sea-
logic evidence. water is also possible. Postore or waning-stage hydrothermal
Quartz-epidote veins located in the hydrothermal cracking activity is supported by the presence of multiple exhalative
zone of the Bell River Complex host primary liquid-vapor- tuffite layers throughout the Wabassee Group and alteration
halite inclusions. These inclusions are interpreted to be sam- signatures extending from the VMS ore deposits into the
ples of the deep-seated equivalent to the ore fluid described hanging wall (Carr, 2004; Ioannou, 2004). Ore-hosted sec-
above. Microthermometry indicates that these inclusions ondary liquid-vapor fluid inclusions have similar temperature
trapped a high-temperature brine (Tt = 373° ± 44°C, 1σ, n = and high-salinity characteristics (Th = 114° ± 14°C and 23.9 ±
92; 38.2 ± 1.9 wt % NaCl equiv, 1σ, n = 92). We interpret this 1.8 wt % NaCl-CaCl2 equiv, 1σ, n = 43), suggesting a similar
brine to be a product of (modified) phase-separated seawater origin.
(3.2 wt % NaCl), an exsolved magmatic component, or a com- Hydrothermal activity ~32 m.y. younger than the VMS sys-
bination, deep within the hydrothermal system at 650° to tem includes quartz veins interpreted to be associated with
670°C and at a near-lithostatic pressure of 90 MPa. Phase the Olga Lake granitoid pluton. Microthermometry of pri-
separation and subsequent convection lowered the tempera- mary liquid-vapor inclusions hosted by the veins indicates
ture of the brine prior to its entrapment within the hy- that the fluids were of higher temperature (Th = 340° ± 20°C,
drothermal cracking zone. The occurrence of the high-tem- 1σ, n = 24), reflecting the thermal influence of the pluton,
perature brine overlain by lower temperature and/or salinity and of moderate salinity (8.0 ± 0.6 wt % NaCl equiv, 1σ, n =
fluid suggests a two-cell convection model, consistent with 24).
metal mass-balance calculations for the south limb (Ioannou, This fluid inclusion data illustrate the importance of sub-
2004). sea-floor chemical and physical processes directly related to
The high salinity of the ore-hosted fluid inclusions is com- metal transport and deposition in VMS systems. In particular,
patible with two possible models. In one model, a significant phase separation deep within the hydrothermal system is in-
amount of brine was incorporated into the upper cell and terpreted as a key process for generating saline brines capa-
mixed with heated seawater during convection. Assuming an ble of forming significant ore deposits.
Archean seawater salinity of 3.2 wt percent NaCl, the relative
amount of brine averaged ~0.6 kg per kg of seawater in the Acknowledgments
upper cell. The resultant hydrothermal fluid rose toward the We thank Grant Arnold and Noranda Inc. for their logis-
sea floor to produce the VMS deposits and, in some places, tical support during the acquisition of field data from the
becomes trapped as primary inclusions. In another model, Matagami district. Colin Bray, F.G. Smith Fluid Inclusion
Archean seawater itself was very saline and of variable salin- Laboratory, University of Toronto, provided expertise and
ity (e.g., de Ronde et al., 1997a). The microthermometrically insight during the microthermometric study. Funding for
indicated high CaCl2 content (max XNaCl <0.55, molar Na/Ca this research was provided by an NSERC (file 661-069-97)
= 2.3/1) of the ore fluid is consistent with basalt-seawater in- and a PGS-B scholarship, CAMIRO (project 96E01), the
teraction (e.g., Hardie, 1983) and an Archean seawater com- Ontario Government Graduate Scholarship program (OGS
position more Ca rich than its present-day equivalent. and OGSST), and the Department of Geology, University of
With the cooling of the Bell River Complex, lower temper- Toronto (Steven D. Scott, Chair). Economic Geology re-
ature fluids, dominantly of seawater origin, circulated deep viewers David Cooke, Phil Brown, Jan Peter, and Mark
within the hydrothermal system. Modified by water-rock in- Hannington provided constructive input that benefited this
teraction, yet not phase separated, these fluids sealed off the manuscript.
remnant permeability of the complex fracture network in the
hydrothermal cracking zone and were locally trapped as sec- February 20, 2004; July 18, 2007
ondary liquid-vapor fluid inclusions of moderate to high salin-
ity (Th = 242° ± 17°C and 9.1 ± 1.6 wt % NaCl equiv, 1σ, n = REFERENCES
14) hosted within quartz-epidote vein material in the Bell Barnes, H.L., 1965, Environmental limitations to mechanisms of ore trans-
River Complex. port: Geological Survey of Czechoslovakia, Problems of Postmagmatic Ore
Deposition Symposium, Prague, v. 2, p. 316–326.
Postore or waning-stage hydrothermal activity at Matagami ——1979, Solubility of ore minerals, in Barnes, H.L., ed., Geochemistry of
is evident from the presence of quartz-epidote veins cross- hydrothermal ore deposits, 2nd ed.: New York, Wiley, p. 404–460.
cutting the hanging-wall Wabassee Group basalts. Microther- Beaudry, C., and Gaucher, E., 1986, Cartographie géologique dans la région
mometry on quartz-hosted primary liquid-vapor fluid include Matagami: Québec Ministère de l’Energie et des Ressources Report
sions suggests that this activity occurred at relatively low MB 86–32, 147 p.
Berndt, M.E., Seyfried, W.E., and Beck, J.W., 1988, Hydrothermal alteration
temperatures (Th = 76°–177°C, n = 212) over a wide range of processes at midocean ridges: Experimental and theoretical constraints
fluid salinity (6.0–32.4 wt % NaCl-CaCl2 equiv, n = 212) and from Ca and Sr exchange reactions and SR isotopic ratios: Journal of Geo-
with a high apparent CaCl2 content (max XNaCl <0.06). The physical Research, v. 93, p. 4573–4583.
0361-0128/98/000/000-00 $6.00 710

Bischoff, J.L., 1991, Densities of liquids and vapors in boiling NaCl-H2O so- Channer, D.M.D., Bray, C.J., and Spooner, E.T.C., 1999, Integrated cation-
lutions: A PVTX summary from 300° to 500°C: American Journal of Sci- anion/volatile fluid inclusion analysis by gas and ion chromatography:
ence, v. 291, p. 309–338. Methodology and examples: Chemical Geology, v. 154, p. 59–82.
Bischoff, J.L., and Pitzer, K.S., 1985, Phase relations and adiabats in boiling Chou, I., 1987, Phase relations in the system NaCl-KCl-H2O. III. Solubilities
seafloor geothermal systems: Earth and Planetary Science Letters, v. 68, p. of halite in vapor-saturated liquids above 445°C and redetermination of
172–180. phase equilibrium properties in the system NaCl-H2O to 1000°C and 1500
Bischoff, J.L., and Rosenbauer, R.J., 1985, An empirical equation of state for bars: Geochimica et Cosmochimica Acta, v. 51, p. 1965–1975.
hydrothermal seawater: American Journal of Science, v. 285, p. 725–763. Cloete, M., 1994, Aspects of the volcanism and metamorphism of the On-
——1988, Liquid-vapor relations in the critical region of the system NaCl- verwacht Group lavas in the southwestern portion of the Barberton green-
H2O from 380° to 415°C: A refined determination of the critical point and stone belt: Unpublished Ph.D. dissertation, Johannesburg, University of
two-phase boundary of seawater: Geochimica et Cosmochimica Acta, v. 52, Witwatersrand, 232 p.
p. 2121–2126. ——1999, Aspects of the volcanism and metamorphism of the Onverwacht
——1989, Salinity variations in submarine hydrothermal systems by layered Group lavas in the southwestern portion of the Barberton greenstone belt:
double-diffusive convection: Journal of Geology, v. 97, p. 613–623. South Africa Council for Geosciences, Memoir 84, 232 p.
Bischoff, J.L, and Seyfried, W.E., 1978, Hydrothermal chemistry of seawater Cloke, P.L., and Kesler, S., 1979, The halite trend in hydrothermal solutions:
from 25° to 350°C: American Journal of Science, v. 278, p. 838–860. ECONOMIC GEOLOGY, v. 74, p. 1823–1831.
Bodnar, R.J., 1992, The system H2O-NaCl [abs.]: Biennial PACROFI, 4th, Clynne, M.A., Potter, R.W., and Haas, J.L., 1981, Solubility of NaCl in aque-
Lake Arrowhead, California, May 22–24, 1992, Program and Abstracts, p. ous electrolyte solutions from 10 to 100°C: Journal of Chemical Engineer-
108–111. ing Data, v. 26, p. 395–398.
——1993, Revised equation and table for determining the freezing point de- Copeland, C.S., Silverman, J.S., and Benson, S.W., 1953, The system NaCl-
pression of H2O-NaCl solutions: Geochimica et Cosmochimica Acta, v. 58, H2O at supercritical temperatures and pressures: Journal of Physical
p. 1053–1063. Chemistry, v. 21, p. 12–16.
——1994, Synthetic fluid inclusions: XII. The system H2O-NaCl. Experi- Costa, U.R., 1984, Hydrothermal footwall alteration and ore formation at the
mental determination of the halite liquidus and isochores for a 40 wt % Mattagami Lake mine, Mattagami, Quebec: Unpublished Ph.D. thesis,
NaCl solution: Geochimica et Cosmochimica Acta, v. 58, p. 1053–1063. London, Ontario, University of Western Ontario, 288 p.
Bodnar, R.J., and Sterner, S.M., 1985, Synthetic fluid inclusions in natural Costa, U.R., Fyfe, W.S., Kerrich, R., and Nesbitt, H.W., 1980, Archean hy-
quartz. II. Applications to PVT studies: Geochimica et Cosmochimica Acta, drothermal talc evidence for high ocean temperatures: Chemical Geology,
v. 49, p. 1855–1859. v. 30, p. 341–349.
Bodnar, R.J., and Vityk, M.O., 1994, Interpretation of microthermometric Costa, U.R., Barnett, R.L., and Kerrich, R., 1983, The Mattagami Lake
data for H2O-NaCl fluid inclusions, in Vivo, B.D., and Frezzotti, M.L., eds., mine Archean Zn-Cu sulfide deposit, Quebec: Hydrothermal coprecipi-
Fluid inclusions in minerals: Methods and applications: International Min- tation of talc and sulfides in a seafloor brine pool—evidence from geo-
eralogical Association Short Course Manual, p. 117–130. chemistry, 18O/16O, and mineral chemistry: ECONOMIC GEOLOGY, v. 78, p.
Borisenko, A.S., 1977, Study of the salt composition of solutions of gas-liquid 1144–1203.
inclusions in minerals by the cryometric method: Soviet Geology and Geo- ——1984, The Mattagami Lake mine Archean Zn-Cu sulfide deposit, Que-
physics, v. 18, p. 11–19. bec: Hydrothermal coprecipitation of talc and sulfides in a seafloor brine
Bourcier, W.L., and Barnes, H.L., 1987, Ore solution chemistry-VII. Stabili- pool—evidence from geochemistry, 18O/16O, and mineral chemistry—a
ties of chloride and bisulfide complexes of zinc to 350°C: ECONOMIC GE- reply: ECONOMIC GEOLOGY, v. 79, p. 1953–1955.
OLOGY, v. 82, p. 1839–1863. Cowan, J., and Cann, J.R., 1988, Supercritical two-phase separation of hy-
Bowers, T.S., Campbell, A.C., Measures, C.I., Spivack, A.J., Khadem, M., drothermal fluids in the Troodos ophiolite: Nature, v. 333, p. 259–261.
and Edmond, J.M., 1988, Chemical controls on the composition of vent flu- Crerar, D.A., and Barnes, H.L., 1976, Ore solution chemistry V. Solubilities
ids at 13°–11°N and 21°N, East Pacific Rise: Journal of Geophysical Re- of chalcopyrite and chalcocite assemblages in hydrothermal solution at
search, v. 93, p. 4522–4536. 200° to 350°C: ECONOMIC GEOLOGY, v. 71, p. 772–794.
Bray, C.J., Spooner, E.T.C., and Thomas, A.V., 1991, Fluid inclusion volatile Davidson, A.J., 1977, Petrography and chemistry of the Key Tuffite at Bell
analysis by heated crushing, on-line gas chromatography: Applications to Allard, Matagami, Quebec: Unpublished M.Sc. thesis, Montreal, Quebec,
Archean fluids: Journal of Geochemical Exploration, v. 42, p. 167–193. McGill University, 83 p.
Burnham C.W., 1997, Magmas and hydrothermal fluids, in Barnes, H.L., ed., Davis, D.W., Lowenstein, T.K., and Spencer, R.J., 1990, Melting behaviour
Geochemistry of hydrothermal ore deposits, 3rd ed.: New York, Wiley, p. of fluid inclusions in laboratory-grown halite crystals in the systems NaCl-
63–123. H2O, NaCl-KCl-H2O, NaCl-MgCl2–H2O and NaCl-CaCl2–H2O: Geochim-
Butterfield, D.A., and Massoth, G.J., 1994, Geochemistry of north Cleft seg- ica et Cosmochimica Acta, v. 54, p. 591–601.
ment vent fluids: Temporal changes in chlorinity and their possible rela- de Ronde, C.E.J., 1995, Fluid chemistry and isotopic characteristics of
tions to recent volcanism: Journal of Geophysical Research, v. 99, p. seafloor hydrothermal systems and associated VMS deposits: Potential for
4951–4968. magmatic contributions: Mineralogical Association of Canada Short
Butterfield, D.A., Massoth, G.J., McDuff, R.E., Lupton, J.E., and Lilley, Course, v. 23, p. 479–510.
M.D., 1990, Geochemistry of hydrothermal fluids from Axial Seamount hy- de Ronde, C.E.J., Channer, D.M.D., Faure, K., Bray, C.J., and Spooner,
drothermal emissions study vent field, Juan de Fuca Ridge: Subsea floor E.T.C., 1997a, Fluid chemistry of Archean seafloor hydrothermal vents:
boiling and subsequent fluid-rock interactions: Journal of Geophysical Re- Implications for the composition of circa 3.2 Ga seawater: Geochimica et
search, v. 95, p. 12895–12921. Cosmochimica Acta, v. 61, p. 4025–4042.
Campbell, I.H., Franklin, J.M., Gorton, M.P., Hart, T.R., and Scott, S.D., de Ronde, C.E.J., Channer, D.M.D., and Spooner, E.T.C., 1997b, Archean
1981, The role of subvolcanic sills in the generation of massive sulfide de- fluids, in De Wit, M., and Ashwal, L.D., eds., Greenstone belts: New York,
posits: ECONOMIC GEOLOGY, v. 76, p. 2248–2253. Oxford, p.309–335.
Carr, P.M., 2004, Physical and chemical consequences of the cooling of sills de Ronde, C.E.J., Bray, C.J., Spooner, E.T.C., and Channer, D.M.D., 2002,
by pore water convection with application to massive sulfide deposits: Un- Direct measurement of hydrothermal fluids from ancient seafloor vents:
published Ph.D. thesis, Ithaca, New York, Cornell University, 130 p. Implications for the composition of Archean seawater [abs.]: Geological So-
Cathles, L.M., 1983, An analysis of the hydrothermal system responsible for ciety of America Abstracts with Programs, v. 34, p. 516.
massive sulfide deposition in the Hokuroku basin of Japan: ECONOMIC GE- Delaney, J.R., 1982, Generation of high salinity fluids from seawater by two
OLOGY MONOGRAPH 5, p. 439–487. phase separation: EOS Transactions, Transactions of the American Geo-
——1993, A capless 350°C flow zone model to explain megaplumes, salinity physical Union, v. 63, p. 1135.
variations, and high temperature veins in ridge axis hydrothermal systems: Delaney, J.R., and Cosens, B.A., 1982, Boiling and metal deposition in sub-
ECONOMIC GEOLOGY, v. 88, p. 1977–1988. marine hydrothermal systems: Marine Technology Society Journal, v. 16, p.
Channer, D.M.D., de Ronde, C.E.J., and Spooner, E.T.C., 1997, The Cl-Br- 62–66.
I- composition of ~3.23 Ga modified seawater: Implications for the geo- Delaney. J.R., Mogk, D.W., and Mottl, M.J., 1987, Quartz-cemented breccias
logical evolution of ocean halide chemistry: Earth and Planetary Science from the Mid-Atlantic Ridge: Samples of a high-salinity hydrothermal up-
Letters, v. 150, p. 325–335. flow zone: Journal of Geophysical Research, v. 92, p. 9175–9192.
0361-0128/98/000/000-00 $6.00 711

712 IOANNOU ET AL.
Dubois, M., and Marignac, C., 1997, The H2O-NaCl-MgCl2 ternary phase di- Harrigan, D.B., and MacLean, W.H., 1976, Petrography and geochemistry of
agram with special application to fluid inclusion studies: ECONOMIC GEOL- epidote alteration patches in gabbro dykes at Matagami, Quebec: Canadian
OGY, v. 92, p. 114–119. Journal of Earth Sciences, v. 13, p. 500–511.
Edmond, J.M., Measure, C., McDuff, R.E., Chan, L.H., Collier, R., Grant, Haynes, F.M., 1985, Determination of fluid inclusion compositions by se-
B., Gordon, L.I., and Corliss, J.B., 1979, Ridge crest hydrothermal activity quential freezing: ECONOMIC GEOLOGY, v. 80, p. 1436–1439.
and the balances of the major and minor elements in the ocean: the Gala- Hedenquist, J.W., 1984, Adiabatic boiling and ocean metal transport: Nature,
pagos data: Earth and Planetary Science Letters, v. 46, p. 1–18. v. 310, p. 636.
Farr, J.E., 1984, The geology, mineralogy, and geochemistry of the 070 faults Hilbert, R., 1979, PVT-Daten von Wasser und von Wassrigen Natriumchlo-
of the Corbet mine, Noranda, Quebec: Toronto, Ontario, Unpublished rid-Losungen bis 873 K, 4000 bar und 25 GEW-% NaCl: Unpublished
M.Sc. thesis, University of Toronto, 154 p. Ph.D. thesis, Karlsruhe, West Germany, University of Fredericiana, 212 p.
Finlow-Bates, T., 1980, The chemical and physical controls on the genesis of Holland, H.D., 1972, The geologic history of seawater—an attempt to solve
submarine exhalative orebodies and their implications for formulating ex- the problem: Geochimica et Cosmochimica Acta, v. 36, p. 637–651.
ploration concepts. A review: Geologisches Jahrbuch, v. D40, p. 131–168. Huppert, H.E., and Sparks, R.S.J., 1984, Double-diffusive convection due to
Fournier, R.O., 1987, Conceptual models of brine evolution in magmatic hy- crystallization in magmas: Annual Review of Earth and Planetary Sciences,
drothermal systems: U.S. Geological Survey Professional Paper 1350, p. v. 12, p. 11–37.
1487–1506. Huppert, H.E., and Turner, J.S., 1981, Double-diffusive convection: Journal
Fox, C.G., 1990, Consequences of phase separation on the distribution of hy- of Fluid Mechanics, v. 106, p. 299–329.
drothermal fluids at ASHES vent field, Axial volcano, Juan de Fuca Ridge: Hutchinson, R.W., Fyfe, W.S., and Kerrich, R., 1980, Deep fluid penetration
Journal of Geophysical Research, v. 95, p. 12923–12926. and ore deposition: Minerals Science Engineering, v. 12, p. 107–120.
Francheteau, J., Needham, H.D., Choukroune, P., Juteau, T., Seguret, M., Ioannou, S.E., 2004, Hydrothermal fluid chemistry and flow paths associated
Ballard, R.D., Fox, P.J., Normark, W., Carranza, A., Cordoba, A., Guerrero, with the Archean Matagami VMS system, Abitibi subprovince, Canada:
J., Rangin, C., Bougault, H., Cambon, P., and Hekinian, R., 1979, Massive Unpublished Ph.D. thesis, Toronto, Ontario, University of Toronto, 275 p.
deep sea sulphide ore deposits discovered on the East Pacific Rise: Nature, Ioannou, S.E., and Spooner, E.T.C., 2007, Fracture analysis of a volcanogenic
v. 277, p. 523–528. massive sulfide-related hydrothermal cracking zone, upper Bell River
Franklin, J.M., Lydon, J., and Sangster, D.F., 1981, Volcanic-associated mas- Complex, Matagami, Quebec: Application of permeability tensor theory:
sive sulfide deposits: ECONOMIC GEOLOGY 75TH ANNIVERSARY VOLUME, p. ECONOMIC GEOLOGY, v. 102, p. 667–690.
485–627. Ioannou, S.E., Götze, J., Weiershäuser, L., Zubowski, S.M., and Spooner,
Freeman, B.C., 1939, The Bell River Complex, northwestern Quebec: Jour- E.T.C., 2004, Cathodoluminescence characteristics of Archean VMS-re-
nal of Geology, v. 47, p. 27–46. lated quartz: Noranda, Ben Nevis and Matagami districts, Abitibi sub-
Galley, A.R., Jonasson, I.R., and Watkinson, D.H., 2000a, Magnetite-rich province, Canada: Geochemistry, Geophysics, Geosystems, v. 5, Q02007,
calc-silicate alteration in relation to synvolcanic intrusion at the Ansil vol- doi:10.1029/2003GC000613.
canogenic massive sulfide deposit, Rouyn-Noranda, Quebec, Canada: Min- Jolly, W.T., 1978, Metamorphic history of the Archean Abitibi belt: Geologi-
eralium Deposita, v. 35, p. 619–637. cal Survey of Canada Paper 78–10, p. 63–78.
Galley, A.R., van Breeman, O., and Franklin, J., 2000b, The relationship be- Kadko, D., and Moore, W., 1988, Radiochemical constraints on the crustal
tween intrusion-hosted Cu-Mo mineralization and the VMS deposits of the residence time of submarine hydrothermal fluids: Endeavour Ridge:
Archean Sturgeon Lake mining camp, northwestern Ontario: ECONOMIC Geochimica et Cosmochimica Acta, v. 83, p. 53–66.
GEOLOGY, v. 95, p. 1543–1550. Kelley, D.S., 1997, Fluid evolution in slow spreading environments: Proceed-
Gerlach, T.M., 1989, CO2 from magma-chamber degassing: Discussion: Na- ings of the Ocean Drilling Program, Scientific Results, v. 153, p. 399–415.
ture, v. 337, p. 124. Kelley, D.S., and Delaney, J.R., 1987, Two-phase separation and fracturing in
Gibson, H.L., Santaguida, F., Paquette-Mihalasky, F.I., and Watkinson, D.H., mid-ocean ridge gabbros at temperatures greater than 700°C: Earth and
2000, Evolution of regional semi-conformable alteration assemblages Planetary Science Letters, v. 83, p. 53–66.
within an Archean subseafloor hydrothermal system, and relationship to Kelley, D.S., and Robinson, P.T., 1990, Development of a brine-dominated
VMS deposits, at Noranda, Quebec, Canada [abs]: Centre for Ore Deposit hydrothermal system at temperatures of 400–500°C in the upper level se-
Research (CODES), University of Tasmania, Special Publication, v. 3, p. quence, Troodos ophiolite, Cyprus: Geochimica et Cosmochimica Acta, v.
61–63. 54, p. 653–661.
Goldfarb M.S., and Delaney, J.R., 1988, Response of two-phase fluids to frac- Kelley, D.S, Robinson, P.T., and Malpas, J.G., 1992, Processes of brine gen-
ture configurations within submarine hydrothermal systems: Journal of eration and circulation in the oceanic crust: Fluid inclusion evidence from
Geophysical Research, v. 93, p. 4585–4594. the Troodos ophiolite, Cyprus: Journal of Geophysical Research, v. 97, p.
Goldie, R., 1979, Consanguineous Archean intrusive and extrusive rocks, No- 9307–9322.
randa, Quebec: Chemical similarities and differences: Precambrian Re- Kerr, D.J., and Gibson, H.L., 1993, A comparison of the Horne volcanogenic
search, v. 9, p. 275 – 287. massive sulfide deposit and intracauldron deposits of the Mine sequence,
Green, G.R., Solomon, M., and Walshe, J.L., 1981, The formation of the vol- Noranda, Quebec: ECONOMIC GEOLOGY, v. 88, p. 1419–1442.
canic-hosted massive sulfide ore deposit at Rosebery, Tasmania: ECONOMIC Khaibulin, K., and Borisov, N.M., 1966, Experimental investigation of the ther-
GEOLOGY, v. 76, p. 304–338. mal properties of aqueous and vapor solutions of sodium and potassium chlo-
Gregory, R.T., and Taylor, H.P., 1981, An oxygen isotope profile in a section rides at phase equilibrium: Teplofizika Vysokikh Temperatur, v. 4, p. 518–523.
of Cretaceous oceanic crust, Semail Ophiolite, Oman: Evidence for δ18O Knauth, L.P., and Lowe, D.R., 1978, Oxygen isotope geochemistry of cherts
buffering of the oceans by deep (>5 km) seawater-hydrothermal circulation from the Onverwacht Group (3.4 b.y.), Transvaal, South Africa, with impli-
at mid-ocean ridges: Journal of Geophysical Research, v. 86, p. 2737–2755. cations for secular variations in the isotopic composition of cherts: Earth
Griffiths, R.W., 1981, Layered double-diffusive convection in porous media: and Planetary Science Letters, v. 41, p. 209–222.
Journal of Fluid Mechanics, v. 102, p. 221–248. Knott, R., Fouquet, Y., Honnorez, J.H., Petersen, S., and Bohn, M., 1998,
Gutzmer, J., Banks, D.A., Luders, V., Hoefs, J., Beukes, N.J., and von Bezing, Petrology of hydrothermal mineralization: A vertical section through the
K.L., 2003, Ancient sub-seafloor alteration of basaltic andesites of the On- TAG mound: Proceedings of the Ocean Drilling Program, Scientific Re-
geluk Formation, South Africa: Implications for the chemistry of Paleopro- sults, v. 158, p. 5–26.
terozoic seawater: Chemical Geology, v. 201, p. 37–53. Kristmannsdóttir, H., 1979, Alteration of basaltic rocks by hydrothermal ac-
Hall, D.L., Sterner, S.M., and Bodnar, R.J., 1988, Freezing point depression tivity at 100–300°C, in Mortland, M.M., and Farmer, V.C., eds., Interna-
of NaCl-KCl-H2O solutions: ECONOMIC GEOLOGY, v. 83, p. 197–202. tional Clay Conference 1978: Amsterdam, Elsevier Scientific Publishing, p.
Hanor, J.S., 1996, Controls on the solubilization of lead and zinc in basinal 359–367.
brines: Society of Economic Geologists Special Publication 4, p. 483–500. Lacroix, S., Simard, A., Pilote, P., and Dubé, L., 1990, Regional geologic el-
Hardie, L.A., 1983, Origin of CaCl2 brines by basalt-seawater interaction: In- ements and mineral resources of the Harricana-Turgeon belt, Abitibi sub-
sights provided by some simple mass balance calculations: Contributions to province of NW Quebec: Canadian Institute of Mining and Metallurgy
Mineralogy and Petrology, v. 82, p. 205–213. Special Volume 43, p. 313–326.
Harper, G.D., 1999, Structural aspects of hydrothermal venting in Large, R.R., 1992, Australian volcanic-hosted massive sulfide deposits: Fea-
seafloor/ophiolite settings: Reviews in Economic Geology, v. 8, p. 53–73. tures, styles and genetic models: ECONOMIC GEOLOGY, v. 87, p. 471–510.
0361-0128/98/000/000-00 $6.00 712

Liaghat, S., and MacLean, W.H., 1992, The Key Tuffite, Matagami mining Potter, R.W., 1977, Pressure corrections for fluid-inclusion homogenization
district: Origin of the tuff components and mass changes: Exploration and temperatures based on the volumetric properties of the system NaCl-H2O:
Mining Geology, v. 1, p. 197–207. U.S. Geological Survey Journal of Research, v. 6, p. 603–607.
Linke, W.F., 1958, Solubilities of inorganic and metal-organic compounds, Potter, R.W., and Brown, D.L., 1977, The volumetric properties of aqueous
Volume 1: Princeton, NJ, Van Nostrand, 1487 p. sodium chloride solutions from 0° to 500°C at pressures up to 2000 bars
Lister, C.R.B., 1974, On the penetration of water into hot rock: Geophysical based on a regression of available data in the literature: U.S. Geological
Journal of the Royal Astronomical Society, v. 39, p. 465–509. Survey Bulletin 1421–C, 36 p.
——1975, Qualitative theory on the deep end of geothermal systems: Second Reyes, A.G., 1990, Petrology of Philippine geothermal systems and the ap-
United Nations Symposium on the Development and Use of Geothermal plication of alteration mineralogy to their assessment: Journal of Volcanol-
Resources, San Francisco, California, May 20, 1975, p. 459–463. ogy and Geothermal Research, v. 42, p. 279–309.
——1986, Differential thermal stress in the earth: Geophysical Journal of the Richardson, C.J., Cann, J.R., Richards, H.J., and Cowan, J.G., 1987, Metal-
Royal Astronomical Society, v. 86, p. 319–330. depleted root zones of the Troodos ore-forming hydrothermal systems,
Macdonald, A., and Fyfe, W.S., 1983, Serpentinization: A mechanism for mi- Cyprus: Earth and Planetary Science Letters, v. 84, p. 243–253.
croearthquake generation in cooling oceanic crust [abs.]: Geodynamics Roberts, G.R., 1975, The geological setting of the Mattagami Lake mine,
Symposium on the Oceanic Lithosphere-Geodynamics Program, Texas Quebec: A volcanogenic massive sulfide deposit: ECONOMIC GEOLOGY, v.
A&M University, Abstracts, p. 39–40. 70, p. 115–129.
Macdonald, A.J., and Spooner, E.T.C., 1981, Calibration of a Linkam TH 600 Roberts, R.G., and Reardon, E.J., 1978, Alteration and ore-forming
programmable heating-cooling stage for microthermometric examination processes at the Mattagami Lake mine, Quebec: Canadian Journal of Earth
of fluid inclusions: ECONOMIC GEOLOGY, v. 76, p. 1248–1258. Sciences, v. 15, p. 1–21.
MacGeehan, P.J., 1978, The geochemistry of altered volcanic rocks at Roedder, E., 1984, Fluid inclusions: Reviews in Mineralogy, v. 12, 646 p.
Matagami, Quebec: A geothermal model for massive sulphide genesis: Roedder, E., and Bodnar, R.J., 1980, Geologic pressure determinations from
Canadian Journal of Earth Sciences, v. 15, p. 551–570. fluid inclusion studies: Annual Review of Earth and Planetary Sciences, v.
——1979, The petrology and geochemistry of volcanic rocks at Matagami, 8, p. 263–301.
Quebec, and their relationship to massive sulphide mineralization: Unpub- Rosenberg, N.D., Spera, F.J., and Haymon, R.M., 1993, The relationship be-
lished Ph.D. dissertation, Montreal, McGill University, 414 p. tween flow and permeability field in seafloor hydrothermal systems: Earth
MacGeehan, P.J., and MacLean, W.H., 1980a, An Archean sub-seafloor ge- and Planetary Science Letters, v. 116, p. 135–153.
othermal system, “calc-alkali” trends, and massive sulphide genesis: Na- Ruaya, J.R., and Seward, T.M., 1986, The stability of chlorozinc(II) com-
ture, v. 286, p. 767–771. plexes in hydrothermal solutions up to 350°C: Geochimica et Cosmochim-
——1980b, Tholeiitic basalt-rhyolitic magmatism and massive sulphide de- ica Acta, v. 50, p. 651–661.
posits at Matagami, Quebec: Nature, v. 283, p. 153–157. Sato, T., 1977, Kuroko deposits: Their geology, geochemistry, and origin, in
Maier, W.D., Barnes, S.J., and Pellet, T., 1996, The economic significance of Volcanic Processes in Ore Genesis, Institution of Mining and Metallurgy
the Bell River Complex, Abitibi subprovince, Quebec: Canadian Journal of London, p. 153–161.
Earth Sciences, v. 33, p. 967–980. Sawkins, F.J., and Kowalik, J., 1981, The source of ore metals at Buchans:
Michard, G., Alabarède, F., Michard, A., Minster, J.F., Charlou, J.L., and Magmatic versus leaching models: Geological Association of Canada Spe-
Tan, N., 1984, Chemistry of solutions from the 13°N East Pacific Rise cial Paper 22, p. 255–268.
hydrothermal site: Earth and Planetary Science Letters, v. 67, p. Schandl, E.S., and Bleeker, W., 1999, Hydrothermal and metamorphic fluids of
297–307. the Kidd Creek volcanogenic massive sulfide deposit: Preliminary evidence
Mortensen, J.K., 1993a, U-Pb geochronology of the eastern Abitibi sub- from fluid inclusions: ECONOMIC GEOLOGY MONOGRAPH 10, p. 379–387.
province, Part 1: Chibougamau-Matagami-Joutel region: Canadian Journal Schiffman, P., and Smith, B.M., 1988, Petrology and oxygen isotope geo-
of Earth Sciences, v. 30, p. 11–28. chemistry of a fossil seawater hydrothermal system within the Solea graben,
——1993b, U-Pb geochronology of the eastern Abitibi subprovince. Part 2: northern Troodos Ophiolite, Cyprus: Journal of Geophysical Research, v.
Noranda-Kirkland Lake area: Canadian Journal of Earth Sciences, v. 30, p. 93, p. 4612–4624.
29–41. Scott, R.W., 1980, The geology and petrology of a portion of the Bell River
Mottl, M.J., and Holland, H.D., 1978, Chemical exchange during hydrother- Complex in Bourbaux Township, Quebec: Unpublished M.Sc. thesis,
mal alteration of basalt by seawater. I. Experimental results for major and Toronto, Ontario, University of Toronto, 244 p.
minor components of seawater: Geochimica et Cosmochimica Acta, v. 42, Scott, S.D., 1997, Submarine hydrothermal systems and deposits, in Barnes,
p. 1103–1115. H.L., ed., Geochemistry of hydrothermal ore Deposits, 3rd ed.: New York,
Nehlig, P., 1991, Salinity of oceanic hydrothermal fluids: A fluid inclusion Wiley, p. 797–877.
study: Earth and Planetary Science Letters, v. 102, p. 310–325. Seward, T.M., and Barnes, H.L., 1997, Metal transport by hydrothermal ore
——1993, Interactions between magma chambers and hydrothermal sys- fluids, in Barnes, H.L., ed., Geochemistry of hydrothermal ore Deposits,
tems: Oceanic and ophiolitic constraints: Journal of Geophysical Research, 3rd ed.: New York, Wiley, p. 435–486.
v. 98, p. 19621–19633. Seyfried, W.E., 1987, Experimental and theoretical constraints on hy-
Norton, D., and Knight, J., 1977, Transport phenomena in hydrothermal sys- drothermal alteration processes at mid-ocean ridges: Annual Review of
tems: Cooling plutons: American Journal of Science, v. 277, p. 937–981. Earth and Planetary Sciences, v. 15, p. 317–335.
Oakes, C.S., Bodnar, R.J., and Simonson, J.M., 1990, The system NaCl- Seyfried, W.E., and Bischoff, J.L., 1981, Experimental seawater-basalt inter-
CaCl2–H2O: I. the ice liquidus at 1 atm total pressure: Geochimica et Cos- action at 300°C, 500 bars, chemical exchange, secondary mineral formation
mochimica Acta, v. 54, p. 603–610. and implications for the transport of heavy metals: Geochimica et Cos-
Ohmoto, H., and Felder, R.P., 1987, Bacterial activity in the warmer, sulfate- mochimica Acta, v. 45, p. 135–147.
bearing Archean oceans: Nature, v. 328, p. 244–246. Seyfried, W.E., and Mottl, M.J., 1982, Hydrothermal alteration of basalt by
Phillips, G.N., Law, J.D.M., and Myers, R.E., 2001, Is the redox state of the seawater under seawater-dominated conditions: Geochimica et Cos-
Archean atmosphere constrained?: Society of Economic Geologists mochimica Acta, v. 46, p. 985–1002.
Newsletter 47, p. 1–18. Seyfried, W.E., Berndt, M.E., and Janecky, D.R., 1986, Chloride depletions and
Piché, M., Guha, J., Daigneault, R., Sullivan, J.R., and Bouchard, G., 1990, enrichments in seafloor hydrothermal fluids: Constraints from experimental
Les gisements volcanogènes du camp minier de Matagami: structure, basalt alteration studies: Geochimica et Cosmochimica Acta, v. 50, p. 469–475.
stratigraphie et implications métallogeniques: Canadian Institute of Mining Seyfried, W.E., Ding, K., Berndt, M.E., and Chen, X., 1999, Experimental
and Metallurgy Special Volume 43, p. 327–336. and theoretical controls on the composition of mid-ocean ridge hydrother-
Piché, M., Guha, J., and Daigneault, R., 1993, Stratigraphic and structural as- mal fluids: Reviews in Economic Geology, v. 8, p. 181–200.
pects of the volcanic rocks of the Matagami mining camp: Implications for Sharpe, J.I., 1965, Field relations of Matagami sulphide masses bearing on
the Norita ore deposits: ECONOMIC GEOLOGY, v. 88, p. 1542–1559. their disposition in time and space: Canadian Institute of Mining and Met-
Pisutha-Arnond, V., and Ohmoto, H., 1983, Thermal history, and chemical allurgy Bulletin, v. 58, p. 951–964.
and isotopic composition of the ore-forming fluids responsible for the ——1968, Geology and sulfide deposits of the Matagami area, Abitibi-East
Kuroko massive sulfide deposits in the Hokuroku district of Japan: ECO- County: Quebec Department of Natural Resources Geological Report 137,
NOMIC GEOLOGY MONOGRAPH 5, p. 523–558. 122 p.
0361-0128/98/000/000-00 $6.00 713

714 IOANNOU ET AL.
Shepherd, T.J., 1981, Temperature-programmable, heating-freezing stage for Vanko, D.A., Bodnar, R.J., and Sterner, S.M., 1988, Synthetic fluid inclu-
microthermometric analysis of fluid inclusions: ECONOMIC GEOLOGY, v. 76, sions: VIII. Vapor-saturated halite solubility in part of the system NaCl-
p. 1244–1247. CaCl2-H2O with application to fluid inclusions from oceanic hydrothermal
Shepherd, T.J., Rankin, A.H., and Alderton, D.H.M., 1985, A practical guide systems: Geochimica et Cosmochimica Acta, v. 52, p. 2451–2456.
to fluid inclusion studies: Glasgow, Blackie, 239 p. Vanko, D.A., Griffith, J.D., and Erickson, C.L., 1992, Calcium-rich brines
Shinohara, H., 1994, Exsolution of immiscible vapor and liquid phases from and other hydrothermal fluid inclusions from plutonic rocks, Oceanogra-
crystallizing silicate melt: Implications for chlorine and metal transport: pher Transform, Mid-Atlantic Ridge: Geochimica et Cosmochimica Acta, v.
Geochimica et Cosmochimica Acta, v. 58, p. 5215–5221. 56, p. 35–47.
Shosa, J.D., 2000, Overpressure in sedimentary basins: Mechanisms and Vervoort, J.D., White, W.M., Thorpe, R.I., and Franklin, J.M., 1993, Post-
mineralogical implications: Unpublished Ph.D. thesis, Ithaca, New York, magmatic thermal activity in the Abitibi greenstone belt, Noranda and
Cornell University, 234 p. Matagami districts: Evidence from whole-rock Pb isotope data: ECONOMIC
Skinner, B.J, 1966, Thermal expansion: Geological Society of America Mem- GEOLOGY, v. 88, p. 1598–1614.
oir 97, p. 75–96. Von Damm, K.L., 1988, Systematics of and postulated controls on submarine
Solomon, M., Tornos, F., and Gaspar, O.C., 2002, An explanation for many of hydrothermal solution chemistry: Journal of Geophysical Research, v. 93, p.
the usual features of the massive sulphide deposits of the Iberian pyrite 4551–4561.
belt: Geology, v. 30, p. 87–90. Von Damm, K.L., and Bischoff, J.L., 1987, Chemistry of hydrothermal solu-
Solomon, M., Tornos, F., Large, R.R., Badham, J.N.P., Both, R.A., and Zaw, tions from the southern Juan de Fuca Ridge: Journal of Geophysical Re-
K., 2004, Zn-Pb-Cu volcanic-hosted massive sulphide deposits: Criteria for search, v. 92, p. 11334–11346.
distinguishing brine pool-type from black smoker-type sulphide deposition: Von Damm, K.L., Edmond, J.M., Grant, B., Measures, C.I., Waldem, B., and
Ore Geology Reviews, v. 25, p. 259–283. Weiss, R.F., 1985, Chemistry of submarine hydrothermal solutions at 21°N,
Sourirajan, S., and Kennedy, G.C., 1962, The system H2O-NaCl at elevated East Pacific Rise: Geochimica et Cosmochimica Acta, v. 49, p. 2197–2220.
temperatures and pressures: American Journal of Science, v. 260, p. Walker, J.C.G., 1983, Possible limits on the composition of the Archean
115–141. ocean: Nature, v. 302, p. 518–520.
Stakes, D.S., and Vanko, D.A., 1986, Multistage hydrothermal alteration of Weiershäuser, L., and Spooner, E.T.C., 2005, Seafloor hydrothermal fluids,
gabbroic rocks from the failed Mathematician Ridge: Earth and Planetary Ben Nevis area, Abitibi greenstone belt: Implications for Archean (~2.7
Science Letters, v. 79, p. 75–92. Ga) seawater properties: Precambrian Research, v. 138, p. 89–123.
Sterner, S.M., Hall, D.L., and Bodnar, R.J., 1988, Synthetic fluid inclusions. Williams-Jones, A.E., and Samson, I.M., 1990, Theoretical estimation of
V. Solubility relations in the system NaCl-KCl-H2O under vapor-saturated halite in the system NaCl-CaCl2-H2O: Applications to fluid inclusions:
conditions: Geochimica et Cosmochimica Acta, v. 52, p. 989–1005. Canadian Mineralogist, v. 28, p. 299–304.
Stommel, H., Arons, A.B., and Blanchard, D., 1956, An oceanographical cu- Wolfgang, B., Alt, J.C., and Humphris, S.E., 2001, A geochemical and iso-
riosity: the perpetual salt fountain: Deep-Sea Research, v. 3, p. 152–153. topic study of the magmatic-hydrothermal transition in the lower oceanic
Tanger, J.C., and Pitzer, K.S., 1989, Thermodynamics of NaCl-H2O: A new crust (ODP hole 735B) [abs]; Geological Society of America Abstracts with
equation of state for the near-critical region and comparisons with other Programs, v. 33, p. 331.
equations for adjoining regions: Geochimica et Cosmochimica Acta, v. 53, Wood, S.A., and Samson, I.M., 1998, Solubility of ore minerals and complex-
p. 973–987. ation of ore metals in hydrothermal solutions: Reviews in Economic Geol-
Urabe, T., 1987, Kuroko deposit modeling based on magmatic hydrothermal ogy, v. 10, p. 33–80.
theory: Mining Geology, v. 37, p. 159–176. Yang, K., and Scott, S.D., 1996, Possible contributions of a metal-rich mag-
Urabe, T., and Sato, T., 1978, Kuroko deposits of the Kosaka mine, northeast matic fluid to a seafloor hydrothermal system: Nature, v. 383, p. 420–423.
Honshu, Japan—products of submarine hot springs on Miocene seafloor: Yardley, B.W.D., 2005, Metal concentrations in crustal fluids and their rela-
ECONOMIC GEOLOGY, v. 73, p. 161–179. tionship to ore formation: ECONOMIC GEOLOGY, v. 100, p. 613–632.
Urabe, T., Scott, S.D., and Hattori, K., 1983, A comparison of footwall-rock Zhang, Y., and Frantz, J.D., 1987, Determination of the homogenization tem-
alteration and geothermal systems beneath some Japanese and Canadian peratures and densities of supercritical fluids in the system NaCl-KCl-
volcanogenic massive sulfide deposits: ECONOMIC GEOLOGY MONOGRAPH CaCl2-H2O using synthetic fluid inclusions: Chemical Geology, v. 64: p.
5, p. 345–364. 335–350.
Urusova, M.A., 1975, Volume properties of aqueous solutions of sodium Zierenberg, R.A., 1984, The Mattagami Lake mine Archean Zn-Cu sulfide
chloride at elevated temperatures and pressures: Russian Journal of Inor- deposit, Quebec: Hydrothermal coprecipitation of talc and sulfides in a
ganic Chemistry, v. 20, p. 1717–1721. seafloor brine pool —evidence from geochemistry, 18O/16O, and mineral
Vanko, D.A., 1988, Temperature, pressure, and composition of hydrothermal chemistry—a discussion: ECONOMIC GEOLOGY, v. 79, p. 1951–1952.
fluids, with their bearing on the magnitude of tectonic uplift at mid-ocean
ridges, inferred from fluid inclusions in oceanic layer 3 rocks: Journal of
Geophysical Research, v. 93, p. 4595–4611.
0361-0128/98/000/000-00 $6.00 714

APPENDIX
Bell River Complex Salinity and Tt Determination where S = salinity (wt % NaCl).
During microthermometric heating, three-phase primary Conversely, in the case of Th(L-V(L)) prior to Tm(salt) (Fig. A2):
fluid inclusions in quartz-epidote veins from the Bell River
Complex, melting of salt (Tm(salt)) occurred both before and 1. Th(L-V(L)) is measured microthermometrically, and the
after homogenization of the liquid and vapor to the liquid corresponding Ph(L-V(L)) is determined using the three-phase
phase [Th(L-V(L))]. The following figures outline the method curve equation of Bodnar (1992):
used to determine salinity and trapping temperature associ-
ated with these two distinct fluid inclusion homogenization P = –135.99 + 304.37*Ψ – 236.18*Ψ2 +
pathways. In the case of Tm(salt) prior to Th(L-V(L)) (Fig. A1): 78.625*Ψ3 – 10.094*Ψ4 + 0.4244*Ψ5, (A.6)
where Ψ = T°C/100, 180° ≤T°C ≤800°C, P = pressure (bars).
1. Tm(salt) is measured microthermometrically, and salinity
is determined using the equation of Sterner et al. (1988): 2. Tm(salt) is measured microthermometrically, and the cor-
responding pressure at which the isochore intersects the liq-
S (wt % NaCl) = 26.242 + 0.4928*Ψ + 1.42*Ψ2 – uidus (Pl) is approximated using Th(L-V(L)), Ph(L-V(L)), and lin-
0.223Ψ3 + 0.04129*Ψ4 + 0.006295*Ψ5 – early regressed (as a function of Th(L-V(L)) 40 wt percent NaCl
0.001967*Ψ6 + 0.0001112*Ψ7, (A.1) liquid + halite field iso-Th slope values of Bodnar (1994).
where Ψ = T°C/100, 0.1° ≤T°C ≤801°C. 3. Salinity is determined using Tm(salt), Pl, and the slope of
the liquidus (Bodnar, 1994):
2. Th(L-V(L)) is measured microthermometrically, and the
corresponding Ph(L-V(L)) (= Pl) is determined using the two dT/dP (°C/kbar) = –38.38 + 0.90*S – 0.0029*S2, (A.7)
phase curve data of Bischoff (1991). where S = salinity (wt % NaCl).
3. Tt is determined using Th(L-V(L)), Pl, Pt (90 MPa estimate), A computer algorithm is used to calculate the liquidus in P-
and the slope of the salinity dependent liquid field iso-Th line T space at various salinities, using the equations of Sterner et
(Bodnar and Vityk 1994): al. (1988), Bodnar (1992), and Bodnar (1994), and to deter-
dP/dT (bar/°C) = as + bs*Th + cs*Th2, (A.2) mine which liquidus passes through the point defined by
Tm(salt) and Pl (a unique solution defined by salinity).
where 4. Tt is approximated using Tm(salt), Pl, Pt (90 MPa estimate)
as = 18.28 + 1.4413*S + 0.0047241*S2 – and the slope of the 40 wt percent NaCl liquid field iso-Th
0.0024213*S3 + 0.000038064*S4, (A.3) line (Bodnar 1994):
bs = 0.019041 – 1.5268*10-2*S + 5.66012*10-4*S2 – dP/dT (bar/°C) = 25.9 – 0.0364*Th. (A.8)

4.2329*10-6*S3 – 3.0354*10-8*S4, (A.4)
cs= –1.5988*10-4 + 3.6892*10-5*S – 1.9473*10-6*S2 +
4.1674*10-8*S3 – 3.3008*10-10*S4, (A.5)
L+H
L+ H
L+H
L+H
(salinit y < 50
(salinit y < 50
(salinity = 5 0 w t.% NaCl)
(salinity = 5 0 w t.% NaCl)
Tmsalt < ThL-V(L) Tmsalt > ThL-V(L)

(salinit y > 50
(salinit y > 50
L+ H
L+H
wt.% NaCl)
wt.% NaCl)
wt.% NaCl)
wt.% NaCl)
Pt 3 Pt 4
ne 1994)
-Th li
Pressure
Pressure
2,3 L is o Bodnar,
4)
Pl l;
9
Na C
, 19
w t. %
d V ne
(4 0
r an Th li
ityk
4)
dna L iso-
n ar, 1 99
(4 0 w t.% H iso -T h lin e
V V
L+ L+
(Bo
NaCl; Bod
e e
ur v ur v
e c 1) e c 1)
p has f, 199 p has f, 199
o f o f
Tw ischo Tw ischo
L+
(B (B
ve L+H+V ve L+H+V
2 e cur e cur
phas phas
PhL-V(L) = Pl Three nar, 1992
)
Three nar, 1992
)
(Bod (Bod
1 PhL-V(L) 1
schematic - not to scale schematic - not to scale
Tm salt ThL-V(L) Tt ThL-V(L) Tm salt Tt

Temperatur e Temperatur e
FIG. A1 FIG. A2
0361-0128/98/000/000-00 $6.00 715

0361-0128/98/000/000-00 $6.00 716

38 - Ioannou Spooner Barrie 2009 Matagami FL Incl 2007

Uploaded by

Copyright:

Available Formats

38 - Ioannou Spooner Barrie 2009 Matagami FL Incl 2007

Uploaded by

Document Information

Original Description:

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

38 - Ioannou Spooner Barrie 2009 Matagami FL Incl 2007

Uploaded by

Copyright:

Available Formats

©2007 Society of Economic Geologists, Inc.

Economic Geology, v. 102, pp. 691–715

Fluid Temperature and Salinity Characteristics of the

0361-0128/98/000/000-00 $6.00 692

0361-0128/98/000/000-00 $6.00 693

0361-0128/98/000/000-00 $6.00 694

0361-0128/98/000/000-00 $6.00 695

0361-0128/98/000/000-00 $6.00 696

0361-0128/98/000/000-00 $6.00 697

Inclusion host and Th(L-V ± H(L ± H) Density

termined using the data of Usurova (1975)

0361-0128/98/000/000-00 $6.00 698

Salinity (wt % Density

Salinity (wt % Density

0361-0128/98/000/000-00 $6.00 699

summarized in Tables 1 and 2 and Figures 5 and 6, Tm(ice) val-

0361-0128/98/000/000-00 $6.00 700

30 VMS (primary: Tt)

VMS (secondary : Th)

20 BRC (primary: Tt)

BRC (secondary : Th)

0361-0128/98/000/000-00 $6.00 701

0361-0128/98/000/000-00 $6.00 702

0361-0128/98/000/000-00 $6.00 703

3.2 wt.% NaCl

5.3 wt.% NaCl

Depth Below Seafloor (km )

0361-0128/98/000/000-00 $6.00 704

0361-0128/98/000/000-00 $6.00 705

0361-0128/98/000/000-00 $6.00 706

Metal solubility The solubility of Cu as chloride complexes is more tempera-

Copper Solubility (ppm)

0361-0128/98/000/000-00 $6.00 707

0361-0128/98/000/000-00 $6.00 708

0361-0128/98/000/000-00 $6.00 709

0361-0128/98/000/000-00 $6.00 710

0361-0128/98/000/000-00 $6.00 711

0361-0128/98/000/000-00 $6.00 712

0361-0128/98/000/000-00 $6.00 713

0361-0128/98/000/000-00 $6.00 714

bs = 0.019041 – 1.5268*10-2*S + 5.66012*10-4*S2 – dP/dT (bar/°C) = 25.9 – 0.0364*Th. (A.8)

(salinity = 5 0 w t.% NaCl)

Tmsalt < ThL-V(L) Tmsalt > ThL-V(L)

Tm salt ThL-V(L) Tt ThL-V(L) Tm salt Tt

0361-0128/98/000/000-00 $6.00 715

You might also like

bs = 0.019041 – 1.526810-2S + 5.6601210-4S2 – dP/dT (bar/°C) = 25.9 – 0.0364*Th. (A.8)