See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/274368373

Stable Isotope Systematics of Sulfate Minerals

Article in Reviews in Mineralogy and Geochemistry · January 2000

DOI: 10.2138/rmg.2000.40.12

CITATIONS READS
174 3,190

3 authors, including:

Charles N Alpers Robert O Rye

United States Geological Survey United States Geological Survey, Lakewood CO
178 PUBLICATIONS 6,710 CITATIONS 163 PUBLICATIONS 11,042 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Arsenic Geochemistry, Mineralogy, and Bioavailability in Mine Waste View project

Reservoir Characterization View project

All content following this page was uploaded by Charles N Alpers on 11 December 2016.

The user has requested enhancement of the downloaded file.

12 Stable Isotope Systematics of Sulfate Minerals
Robert R. Seal, II
U.S. Geological Survey
954 National Center
Reston, Virginia 20192

Charles N. Alpers
U.S. Geological Survey
6000 J Street
Sacramento, California 95819

Robert O. Rye
U.S. Geological Survey
963 Denver Federal Center
Denver, Colorado 80225

INTRODUCTION
Stable isotope studies of sulfate minerals are especially useful for unraveling the
geochemical history of geological systems. All sulfate minerals can yield sulfur and oxygen
isotope data. Hydrous sulfate minerals, such as gypsum, also yield oxygen and hydrogen
isotope data for the water of hydration, and more complex sulfate minerals, such as alunite
and jarosite also yield oxygen and hydrogen isotope data from hydroxyl sites. Applications
of stable isotope data can be divided into two broad categories: geothermometry and tracer
studies. The equilibrium partitioning of stable isotopes between two substances, such as the
isotopes of sulfur between barite and pyrite, is a function of temperature. Studies can also
use stable isotopes as a tracer to fingerprint various sources of hydrogen, oxygen, and
sulfur, and to identify physical and chemical processes such as evaporation of water,
mixing of waters, and reduction of sulfate to sulfide.
Studies of sulfate minerals range from low-temperature surficial processes associated
with the evaporation of seawater to form evaporite deposits to high-temperature magmatic-
hydrothermal processes associated with the formation of base-and precious-metal deposits.
Studies have been conducted on scales from submicroscopic chemical processes associated
with the weathering of pyrite to global processes affecting the sulfur budget of the oceans.
Sulfate isotope studies provide important information to investigations of energy and
mineral resources, environmental geochemistry, paleoclimates, oceanography (past and
present), sedimentary, igneous, and metamorphic processes, Earth systems,
geomicrobiology, and hydrology.
One of the most important aspects of understanding and interpreting the stable isotope
characteristics of sulfate minerals is the complex interplay between equilibrium and kinetic
chemical and isotopic processes. With few exceptions, sulfate minerals are precipitated
from water or have extensively interacted with water at some time in their history. Because
of this nearly ubiquitous association with water, the kinetics of isotopic exchange reactions
among dissolved species and solids are fundamental in dictating the isotopic composition
of sulfate minerals. In general, the heavier isotope of sulfur is enriched in the higher
oxidation state, such that under equilibrium conditions, sulfate minerals (e.g. barite,
anhydrite) are expected to be enriched in the heavy isotope relative to disulfide minerals
(e.g. pyrite, marcasite), which in turn are expected to be enriched relative to monosulfide
minerals (e.g. pyrrhotite, sphalerite, galena) (Sakai 1968, Bachinski 1969). The kinetics of
1529-6466/00/0040-0012$10.00 DOI: 10.2138/rmg.2000.40.12
542 Seal, Alpers & Rye

isotopic exchange among minerals with sulfur at the same oxidation state, such as
sphalerite, and galena, are such that equilibrium is commonly observed. In contrast,
isotopic equilibrium for exchange reactions between minerals of different oxidation states
depends on factors such as the pH, time and temperature of reaction, the direction of
reaction, fluid composition, and the presence or absence of catalysts (Ohmoto and Lasaga
1982). The kinetics of oxygen isotope exchange between dissolved sulfate and water are
extremely sluggish. Extrapolation of the high-temperature (100 to 300°C) isotopic
exchange kinetic data of Chiba and Sakai (1985) to ambient temperatures suggests that it
would take several billions of years for dissolved sulfate and seawater to reach oxygen
isotopic equilibrium. In contrast, the residence time of sulfate in the oceans is only 7.9
million years (Holland 1978). However, at higher temperatures (>200°C), oxygen isotopic
exchange is sufficiently rapid to permit application of sulfate isotope geothermometry to
geothermal systems and hydrothermal mineral deposits. In general, equilibrium prevails at
low pH and high temperatures, whereas kinetic factors preclude equilibrium at low
temperatures even at low pH. Thus, the sluggish kinetics of sulfur and oxygen isotope
exchange reaction at low temperatures impair the use of these isotopes to understand the
conditions of formation of sulfate minerals in these environments. However, because of
these slow kinetics, the oxygen and sulfur isotopic compositions of sulfate minerals may
preserve a record of the sources and processes that initially produced the dissolved sulfate,
because the isotope ratios may not re-equilibrate during fluid transport and mineral
precipitation.
The first part of this chapter is designed to provide the reader with a basic
understanding of the principles that form the foundations of stable isotope geochemistry.
Next, an overview of analytical methods used to determine the stable isotope composition
of sulfate minerals is presented. This overview is followed by a discussion of
geochemical processes that determine the stable isotope characteristics of sulfate minerals
and related compounds. The chapter then concludes with an examination of the stable
isotope systematics of sulfate minerals in a variety of geochemical environments.

FUNDAMENTAL ASPECTS OF STABLE ISOTOPE GEOCHEMISTRY

An isotope of an element is defined by the total number of protons (Z) and neutrons
(N) present, which sum together to give the atomic mass (A). For example, the element
oxygen, abbreviated as “O”, is defined by the presence of 8 protons, but can have either
8, 9, or 10 neutrons. These various combinations of 8 protons and either 8, 9, or 10
neutrons are the isotopes of oxygen and are denoted as oxygen-16 (8Z + 8N = 16A),
oxygen-17 (8Z + 9N = 17A), and oxygen-18 (8Z + 10N =18A). The isotopes of oxygen
are abbreviated as 16O, 17O, and 18O, respectively. For oxygen, all of these isotopes are
stable, meaning that they do not undergo radioactive decay at significant rates. In
contrast, carbon (Z = 6) has two stable isotopes, 12C and 13C, and one unstable isotope,
14
C. Unstable, or radiogenic isotopes undergo radioactive decay. For example, 14C decays
to 14N, a stable isotope. The decay of radiogenic isotopes makes many of them useful for
dating purposes. This chapter is concerned only with naturally occurring stable isotopes.
Sulfur (Z = 16) has four stable isotopes: 32S, 33S, 34S, and 36S, with approximate
terrestrial abundances of 95.02, 0.75, 4.21, and 0.02 percent, respectively (Macnamara
and Thode 1950). Oxygen (Z = 8) has three stable isotopes: 16O, 17O, and 18O, with
approximate terrestrial abundances of 99.763, 0.0375, and 0.1995 percent, respectively
(Garlick 1969). Hydrogen (Z = 1) has two stable isotopes: 1H, and 2H, with approximate
terrestrial abundances of 99.9844, and 0.0156 percent, respectively (Way et al. 1950).
Deuterium, 2H, is commonly abbreviated as “D.” In this context, 1H is further abbreviated
merely as “H.”
 Stable Isotope Systematics of Sulfate Minerals 543

Stable isotope geochemistry is concerned primarily with the relative partitioning of

stable isotopes among substances (i.e. changes in the ratios of isotopes), rather than their
absolute abundances. Differences in the partitioning of stable isotopes, otherwise known as
fractionation, are due to equilibrium and kinetic effects. In general, heavier isotopes form
more stable bonds; molecules of different masses react at different rates (O’Neil 1986).
Stable isotope geochemists investigate the variations of a minor isotope of an element
relative to a major isotope of the element. For sulfate minerals, the principal ratios of
concern are D/H, 18O/16O, and 34S/32S. For oxygen, the ratio 18O/16O is approximately
0.00204. Fractionation processes will typically cause variations in this ratio in the fifth or
sixth decimal places. Because we are concerned with variations in isotopic ratios that are
relatively small, the isotopic composition of substances is expressed in the “delta” (δ)
notation as permil (or parts per thousand) variation relative to a reference material. The δ-
notation for the 18O/16O composition of a substance is defined as:
⎛
δ18O⎜
(
18
) ( O/ O)
O/16 O sample − 18 16
reference
⎞
⎟ ×1000 (1)
⎜
⎝ ( O/ O)
18 16
reference
⎟
⎠
expressed in values of parts per thousand or permil (‰), which can also be found in the
literature spelled “per mil”, “per mill”, and “per mille”. For most applications, the agreed
upon reference for oxygen isotopes is Vienna Standard Mean Ocean Water (VSMOW),
for which δ18O = 0.0 ‰ by definition. VSMOW is a hypothetical water with oxygen and
hydrogen isotopic compositions similar to those of average ocean water. A sample with a
positive δ18OVSMOW value, such as +5.0 ‰, is enriched in δ18O relative to VSMOW.
Conversely, a sample with a negative δ18OVSMOW value, such as -5.0 ‰, is depleted in
18
O relative to VSMOW. The absolute 18O/16O ratio for VSMOW is 2005.2±0.45 × 10-6
(Baertschi 1976). The Vienna Peedee Belemnite (VPDB) reference is only used for
oxygen isotope values in carbonates, with application to studies of oceanic
paleotemperature and sedimentary carbonate petrology. The VPDB reference was
originally defined by a belemnite sample from the Peedee Formation in North Carolina.
Hydrogen isotopes are also defined relative to VSMOW, using the D/H ratio, such
that the δD value for VSMOW is 0.0 ‰ by definition. The absolute D/H ratio for
VSMOW is 155.76±0.05 × 10-6 (Hageman et al. 1970). The selection of seawater
(VSMOW) as the reference for hydrogen isotope and most oxygen isotope applications in
geochemistry is useful because the oceans are the foundation of the hydrological cycle.
For sulfur isotopes, the δ34S value for the 34S/32S ratio is defined relative to Vienna
Cañon Diablo Troilite (VCDT) with δ34S = 0.0 ‰ by definition. The reference was
originally defined by the isotopic composition of troilite (FeS) from the Cañon Diablo
iron meteorite. The absolute 34S/32S ratio for Cañon Diablo troilite is 4500.45 x 10-6 (Ault
and Jensen 1963). The selection of a meteoritic sulfide mineral as the reference for sulfur
is useful because meteoritic sulfide is thought to represent the primordial sulfur isotopic
composition of Earth (Nielsen et al. 1991). Thus, any variations in the isotopic
composition of terrestrial sulfur relative to VCDT reflect differentiation since the
formation of Earth.
For oxygen and sulfur, which have more than two stable isotopes, 18O/16O and 34S/32S
are the ratios that are almost exclusively measured in studies of terrestrial systems. These
ratios are chosen for two main reasons. First, they represent the most abundant isotopes of
these elements, which facilitates their ease of analysis. Second, isotopic fractionation is
governed by mass balance such that different isotopic ratios tend to vary systematically
with one another in proportions that can be approximated by the mass differences between
the isotopic ratios for closed, terrestrial systems being affected by chemical and physical
544 Seal, Alpers & Rye

processes. In other words, the variations in the 17O/16O ratio of a sample will be
approximately half that of the 18O/16O ratio because of the relative differences in masses.
This linear fractionation trend due to physical and chemical processes is called a “terrestrial
fractionation line” (Clayton 1986). Except as recently shown by Thiemens (1999a,b) and
mentioned below, isotopic data from all samples of terrestrial origins conform to terrestrial
fractionation systematics. In contrast, most meteorites plot away from the terrestrial
fractionation line in terms of δ17O and δ18O; this deviation has been attributed to the
different pathways of nucleosynthesis of the different oxygen isotopes and their subsequent
and heterogeneous contributions to the solar nebula (Clayton 1986). All sulfide minerals
from terrestrial samples and meteorites fall on a single mass-fractionation line suggesting
that any heterogeneities that may have existed were homogenized early in solar history
(Hulston and Thode 1965a,b). The metallic phase of iron meteorites shows anomalies in its
33
S and 36S compositions, which have been attributed to cosmic-ray spallation (Hulston and
Thode 1965b). Recent work by Thiemens (1999a,b) has shown that atmospheric sulfate
aerosols have isotopic compositions that reflect mass-independent oxygen and sulfur
isotopic fractionation. These anomalies have been recognized in the sediment record.
The partitioning of stable isotopes between two substances, A and B, is
quantitatively described by a fractionation factor which is defined as
RA
αA−B = (2)
RB
18
where R is either D/H, O/16O, or 34
S/32S. This equation can be recast in terms of δ
values as
δA
1+
α A −B = 1000 = 1000 + δ A (3)
δ 1000 + δ B
1+ B
1000
Values of α are typically near unity, with variations normally in the third decimal
place (1.00x). For example, the 18O/16O fractionation between anhydrite and water at
300°C yields an α anhydrite−H2O value of 1.0090. Thus, anhydrite is enriched in 18O relative to
water by 9.0 permil (i.e. the fractionation equals + 9.0 ‰). For an α value less than unity,
such as α H 2 O − anhydrite which equals 0.991, the water is depleted in 18O relative to anhydrite
by 9.0 permil (i.e. the fractionation equals -9.0 ‰). In the literature, fractionation factors
may be expressed in a variety of ways including α, 1000 1nα, and Δ, among others. The
value Δ A − B is defined as
Δ A − B = δA −δB (4)
A convenient mathematical relationship is that 1000 1n(1.00X) is approximately equal to
X, so that
Δ A−B ≈ 1000 1nα A−B (5)
Isotopic fractionations may also be defined in terms of an enrichment factor (ε) where
ε A−B ≈ (α A−B − 1)× 1000 (6)
ANALYTICAL METHODS
Several procedures are available to determine the oxygen, sulfur, and hydrogen
isotopic compositions of sulfate minerals. Conventional analyses typically involve mineral-
separation procedures that may include hand-picking or gravimetric techniques (heavy
liquids, panning, etc.) and (or) wet-chemical techniques. Once a suitable concentration of
 Stable Isotope Systematics of Sulfate Minerals 545

the desired compound is obtained, the element of interest (O, S, or H) is extracted and
converted to a gaseous form that is amenable to mass-spectrometric analysis. For oxygen
isotopes, the traditionally preferred gas for analysis is CO2; for sulfur it is SO2; and for
hydrogen it is H2. The amount of sample required varies among laboratories but typically
ranges from 5 to 20 mg of pure mineral separate for δ18O and δ34S, and up to several
hundred mg for δD using conventional techniques. Typical analytical uncertainties (1σ) for
conventional techniques are ±0.1 ‰ for δ18O and δ34S and ± 1‰ for δD.
Sulfate for oxygen and sulfur isotope analysis from simple sulfate minerals, such as
barite, anhydrite, gypsum, and many of the transition metal salts (melanterite,
chalcanthite, etc.) can be purified by reacting with a 5 mass % Na2CO3 solution to leach
sulfate, followed by filtration of the solution, acidification (pH ≈ 4), and reprecipitation
of the sulfate as barite by addition of a 10 mass % BaCl2·2H2O solution. Similarly,
aqueous sulfate can be extracted by filtration, acidification, and precipitation by the
addition of BaCl2·2H2O solution (Taylor et al. 1984). Details of the total isotopic analysis
(δD, δ18OSO 4 , δ18OOH , δ34S) of alunite-group minerals, including mineral separation
techniques, have been described by Wasserman et al. (1992). Field and laboratory
techniques for the extraction of dissolved sulfate and sulfide from waters for isotopic
analysis have been summarized by Carmody et al. (1998).
For conventional analysis of δ18OSO 4 , the CO2 is generally prepared by one of two
methods. In the first method, BaSO4 is mixed with graphite in a platinum-foil boat
through which a current is passed under vacuum; CO2 is produced directly and any CO
produced is converted to CO2 by plating out excess carbon onto electroplates (Rafter
1967, Sakai and Krouse 1971). In the second method, sulfate minerals are reacted with
BrF5 or ClF3 at elevated temperatures (550 to 650°C) to produce O2, which is converted
to CO2 for isotopic analysis by reaction with a hot graphite rod under vacuum (Clayton
and Mayeda 1963, Borthwick and Harmon 1982). Because oxygen extraction by
fluorination techniques is not quantitative, presumably due to the production of S-O-F
gases, a kinetic fractionation correction must be applied to CO2 data from this method
(Pickthorn and O’Neil 1985, Wasserman et al. 1992). In hydroxy-sulfate minerals that do
not have water of hydration (e.g. alunite and jarosite), the δ18OOH is calculated by
difference from the bulk δ18O of the complex sulfate mineral, where the bulk δ18O is
determined by fluorination techniques and the δ18OSO 4 is determined on sulfate extracted
from the mineral (Pickthorn and O’Neil 1985, Wasserman et al. 1992).
For conventional δ34S analysis of sulfate minerals, SO2 is produced for analysis by
reacting the sulfate mineral with an oxidant (CuO, Cu2O, or V2O5) and silica glass at
elevated temperatures (1000 to 1200°C) under vacuum (Holt and Engelkemeier 1970,
Haur et al. 1973, Coleman and Moore 1978). Other conventional techniques for the δ34S
and δ18S analysis of sulfate minerals have been summarized by Rees and Holt (1991). For
δD analysis, minerals are dehydrated under vacuum and the water is converted to H2 for
isotopic analysis either by reaction with uranium (Bigeleisen et al. 1952) or zinc
(Coleman et al. 1982, Vennemann and O’Neil 1993).
Isotopic analysis for all of these isotopes is conducted on a gas-source, sector-type,
isotope ratio mass spectrometer. In gas-source mass spectrometers, gas molecules (CO2,
SO2, or H2) are ionized to positively charged particles, such as SO2+, which are
accelerated through a voltage gradient. The ion beam passes through a magnetic field,
which causes separation of various masses such as 64 (32S16O2) and 66 (34Sl6O2,
34 18 16
S O O). In conventional dual-inlet mass spectrometers, a sample gas is measured
alternately with a reference gas. The beam currents are measured in faraday cups and can
be related to the isotopic ratio when the sample and standard gases are compared.
546 Seal, Alpers & Rye

Recent technological advances over the past decade have opened up two new frontiers
in stable isotope analysis of sulfate minerals. One area is the in situ microanalysis of
minerals, and the other is the mass production of data from small samples. For in situ
analysis, a limited amount of sulfur isotope data has been generated from samples of
anhydrite and barite using the ion microprobe or secondary ion mass spectrometer (SIMS)
(McKibben and Riciputi 1998). The ion microprobe bombards a sample with a beam of
charged CS or O. The ion beam causes the sample to be ablated as secondary ionic species,
which are measured by a mass spectrometer. Spatial resolution is on the order of 30 to 60
μm with an analytical uncertainty of ±1 ‰ for sulfur isotope analyses using a SHRIMP,
Sensitive High Mass Resolution Ion Microprobe (McKibben et al. 1996).
The other recent advance is the development of continuous-flow techniques that use
a combination of an elemental analyzer and gas chromatograph for online combustion
and purification of gases that are carried in a He stream directly into the ion source of a
modern mass spectrometer. Continuous-flow systems can measure the sulfur and oxygen
isotopic ratios of sulfate samples in the microgram range, compared to the milligram
range for conventional techniques (Giesemann et al. 1994, Kornex1 et al. 1999). Sample
gases are prepared by on-line peripheral devices such as elemental analyzers that are
capable of processing 50 to 100 samples in a highly automated fashion. Furthermore,
most sulfur isotope measurements can be made without mineral purification, if bulk
sulfur data are all that is desired.
REFERENCE RESERVOIRS
The isotopic compositions of substances (minerals, waters, gases, biological
material) are typically discussed in terms of geochemical reservoirs which may have
served as sources for these constituents. For waters, the main reservoirs are seawater,
meteoric water, and magmatic water. Modern seawater has a fairly uniform isotopic
composition, with δD and δ18O ranging from -7 to +5 ‰, and -1.0 to +0.5 ‰,
respectively, and with mean values close to the composition of VSMOW (δD = 0.0 ‰;
δ18O = 0.0 ‰) (Epstein and Mayeda 1953, Friedman 1953, Craig and Gordon 1965).
Isotopic studies of ancient marine carbonates suggest that the δ18O of seawater in the
early Phanerozoic may have been as low as -9.0 ‰ (Veizer et al. 1997).
The δD and δ18O of meteoric waters, those that originate as precipitation, vary
systematically and predictably (Epstein and Mayeda 1953, Friedman 1953, Craig 1961,
Dansgaard 1964). The linear variation shown by meteoric waters is known as the “global
meteoric water line” (Fig. 1; Craig 1961) and is described by the equation:
δD = 8 δ18O + 10 (7)
Subsequent work, published more than 30 years later (Rozanski et al. 1993) has slightly
refined this empirical relation:
δD = 8.13 δ18O + 10.8 (8)
The linear relationship is the result of the kinetically controlled evaporation of moisture
dominantly from the oceans, and its subsequent isotopic distillation during condensation
and precipitation. The linear relationship is well-described as a Rayleigh distillation
process, discussed in the next section. Numerous factors determine the position of the
isotopic composition of precipitation along the meteoric water line; chief among these are
temperature, which is a function of latitude, altitude, and season, followed by the
“continent effect” and the “rainout effect”. The “continent effect” refers to systematic
decrease of the isotopic composition of precipitation as one moves toward the interior of
continents. The “rainout effect” refers to the systematic decrease of the isotopic
 Stable Isotope Systematics of Sulfate Minerals 547

Figure 1. Plot of δD and δ18O for various

water reservoirs and trends for physical and
chemical processes that can alter the
isotopic composition of water. VSMOW
indicates Vienna Standard Mean Ocean
Water; PMW indicates primary magmatic
water. All isotopic values in permil
(VSMOW).

composition of precipitation with increased amounts of precipitation. Both effects are the
result of fractionation associated with distillation. The temperature effect results in a
continuous increase in δD and δ18O of precipitation from cold to warm climates (Fig. 1).
The temperature dependence of the isotopic composition of meteoric waters has been
used as the basis for numerous paleoclimate studies including studies of ice-sheet
stratigraphy (Dansgaard et al. 1969), ancient soils (Lawrence and Taylor 1971), supergene
clays (Sheppard et al. 1969), cave deposits (Schwarcz et al. 1976), supergene alunites (Rye et
al. 1992, Arehart and O’Neil 1993), and fluid inclusions in hydrothermal mineral deposits
(O’Neil and Silberman 1974, Seal and Rye 1993), to name a few. However, because of the
variety of factors that control the isotopic composition of meteoric water, inferred
compositions of meteoric water cannot be used for quantitative determination of
paleotemperatures. At best, changes in composition reflect relative changes in the
temperature (Dansgaard et al. 1969), and similar compositions of meteoric waters of different
ages reflect grossly similar mean annual temperatures of precipitation (Seal and Rye 1993).
In addition to the factors discussed above, several other factors affect the position of the
meteoric water line. The relative humidity accompanying evaporation from the oceans can
change the “y-intercept”, known as the “deuterium excess factor” (d) of the meteoric water
line resulting in regional meteoric water lines that are parallel to the global meteoric water
line. In arid environments such as the eastern Mediterranean, d values can reach 23 ‰,
whereas in humid environments such as the South Atlantic, d values can drop to less than 6
‰ (Yurtsever and Gat 1981). In terms of ancient conditions, the isotopic composition of
seawater directly affects d. Secondary processes such as evaporation and high-temperature
reactions with rocks can alter the isotopic composition of meteoric waters.
Magmatic water as used here is water that has equilibrated with a magma, regardless
of origin. Compositions of magmatic water are typically calculated rather than measured
from natural samples because no unambiguous natural samples of magmatic water, other
than those found in fluid inclusions have been found (e.g. Rye 1966, Deen et al. 1994).
Accessible, shallow level magmatic-hydrothermal systems are invariably contaminated by
meteoric waters. The isotopic composition of hypothetical magmatic water is calculated on
the basis of the isotopic composition of unaltered magmatic silicate minerals (quartz,
feldspar, muscovite, biotite, and hornblende), experimentally determined mineral-water
fractionation factors, and estimates of magmatic temperatures. Calculated compositions of
primary magmatic waters span the range δD = -40 to -80 ‰, and δ18O = +5.5 to +9.5 ‰
(Sheppard et al. 1969; Fig. 1). Magmatic waters calculated by such means represent an
oversimplification of the complexity of natural processes that can effect stable isotopic
fractionation such as crystallization history and speciation of water in magmas (Taylor
548 Seal, Alpers & Rye

1986, Taylor 1997). Nevertheless, calculated compositions of magmatic water provide a

convenient reference from which to assess the relative importance of various geochemical
processes in determining the isotopic compositions of waters and minerals precipitated
from these processes.
Several common chemical and physical processes can act upon waters to alter their
isotopic composition; chief among these processes are evaporation, water-rock
interaction, and mixing. Evaporation causes an enrichment in D and 18O in the residual
water, thereby producing a positive trend from the starting composition typically with a
slope between 4 and 6 (Fig. 1). Under very saline conditions, however, a different slope
and a hook in the δD-δ18O trajectory may be produced (Sofer and Gat 1975). For dilute
water, the slope is dominantly a function of the relative humidity accompanying
evaporation, but for saline waters, the hydration effect on cations becomes an important
factor. Trends with lower slopes on a δD versus δ18O plot reflect drier conditions and
steeper slopes reflect wetter conditions. Water-rock interactions can alter both the oxygen
and hydrogen isotopic composition of waters. At high water/rock ratios, the effect is
detectable only in the oxygen isotopic composition of the water because of the relatively
small amount of hydrogen relative to oxygen in rocks. The mass ratio of oxygen in water
to oxygen in siliciclastic sedimentary and felsic igneous rocks typically varies around 1.8,
whereas that ratio for hydrogen is around 60. The shape of the interaction paths for the
water in terms of δD and δ18O will vary depending on the starting composition of the
water and rock (Ohmoto and Rye 1974, Criss and Taylor 1986, Seal and Rye 1992).
Geothermal waters commonly have hydrogen isotope compositions that are
indistinguishable from local meteoric waters, except where modified by mixing and
boiling (Truesdell et al. 1977), whereas the oxygen isotope compositions are commonly
enriched in 18O by up to 20 ‰ (Truesdell and Hulston 1980).
An additional reservoir of oxygen contributes to secondary sulfate minerals that
result from the weathering of sulfide minerals. The sulfate oxygen typically comprises
mixtures of oxygen derived from water and oxygen derived from the atmosphere.
Atmospheric oxygen is globally homogeneous, averaging δ18O = 23.5±0.3 ‰ (Dole et al.
1954, Kroopnick and Craig 1972).
For sulfur, the most common reference reservoirs are meteoritic sulfur and seawater.
Meteoritic sulfur, such as Cañon Diablo troilite, provides a convenient reference because
it approximates the Earth’s bulk composition. The iron meteorites have an average sulfur
isotope composition of δ34S = O.2±0.2 ‰ (Kaplan and Hulston 1966) which is indis-
tinguishable from that of pristine mid-ocean ridge basalts (δ34S = 0.3±0.5 ‰; Sakai et al.
1984). Geochemical processes, the most notable of which are oxidation and reduction,
profoundly fractionate sulfur isotopes away from bulk-Earth values in geological systems
(Fig. 2). Oxidation processes produce species that are enriched in 34S relative to the
starting material, whereas reduction produces species that are depleted in 34S.
Oxidation-reduction reactions involving reduced sulfur from the interior of the Earth
throughout its history have resulted in a δ34S of +21.0±0.2 ‰ for dissolved sulfate in
modern oceans (Rees et al. 1978). Because of the volume and importance of the ocean in
the global sulfur cycle, this composition is another important reference reservoir from
which to evaluate sulfur isotope variations in geological systems. The δ34S of sulfate in
ancient oceans as recorded by marine evaporite sequences (Claypool et al. 1980) has
varied from a low of approximately 10 ‰ during Permian and Triassic time to a high of
35 ‰ during Cambrian time. The causes and implications of the secular variations in the
sulfur isotope composition of seawater are discussed in a later section.
 Stable Isotope Systematics of Sulfate Minerals 549

Figure 2. δ34S of various geologic reservoirs. Modified from Krouse (1980).

Dimethylsulfide data from Calhoun et al. (1991); isotopic values in permil (VCDT).

FACTORS THAT CONTROL STABLE ISOTOPE FRACTIONATION

Isotopic fractionation is a result of variations in thermodynamic properties of
molecules that are dependent on mass. Details of the thermodynamic basis for
understanding isotope fractionation have been presented by Urey (1947), Bigeleisen and
Mayer (1947), and Bigeleisen (1952). Isotope fractionation results from equilibrium and
kinetically controlled chemical and physical processes. Equilibrium processes include
isotopic exchange reactions, which redistribute isotopes among molecules of different
substances. Kinetic processes include irreversible chemical reactions and physical
processes such as evaporation and diffusion (O’Neil 1986). Equilibrium isotope effects
result from the effect of atomic mass on bonding; molecules containing a heavier isotope
are more stable than those containing a lighter isotope. Kinetic isotope effects are related to
greater translational and vibrational velocities associated with lighter isotopes. It is easier to
break bonds with lighter isotopes, for example the 32S-O bond, compared with those with
heavier isotopes, such as the 34S-O bond, during the bacterially mediated reduction of
dissolved sulfate to sulfide.
Among the several factors that influence the magnitude of equilibrium stable isotope
fractionations are temperature, chemical composition, crystal structure, and pressure (O’Neil
1986). For the present discussion, temperature and chemical composition are the most
important. Polymorphism is rare in the sulfate minerals discussed here and pressure effects
are minimal at upper crustal conditions. The temperature dependence of fractionation factors
results from the relative effect of temperature on the vibrational energies of two substances.
Theoretical considerations indicate that the stable isotope fractionation between two
substances should approach zero at infinite temperature (Bigeleisen and Mayer 1947). These
fractionations are generally described well by equations of the form:
A B
10001nα = + +C (9)
T2 T
where A, B, and C are empirically determined constants.
The dependence of isotopic fractionation on chemical composition includes such
compositional variables as oxidation state, ionic charge, atomic mass, and the electronic
configuration of the isotopic elements and the elements to which they are bound (O’Neil
1986). For sulfur-bearing systems, the effect of the oxidation state of sulfur is especially
important. The higher oxidation states of sulfur are enriched in the heavier isotopes relative
to lower oxidation states such that 34S enrichment follows the general trend SO42- > SO32- >
550 Seal, Alpers & Rye

Sx0 > S2- (Sakai 1968, Bachinski 1969). In the geological record, this trend is reflected by
the fact that sulfate minerals, such as barite or anhydrite, typically have higher δ34S values
than cogenetic sulfide minerals in a variety of geochemical settings. Cationic substitutions
also play an important role in stable isotope fractionations. O’Neil et al. (1969) documented
a cation-mass dependence of 18O enrichment in divalent metal-carbonate minerals with 18O
enrichment following the order CaCO3 > SrCO3 > BaCO3. Likewise, the 18O enrichment in
divalent sulfate minerals is such that CaSO4 > BaSO4 (Lloyd 1968, Kusakabe and
Robinson 1977, Chiba et al. 1981), and the 34S enrichment follows the order CaSO4·2H2O
> BaSO4 (Thode and Monster 1965, Kusakabe and Robinson 1977). Similarly, for Al3+-
Fe3+ substitution in the grossular-andradite garnets, Taylor and O’Neil (1977) found that
18
O was enriched in natural Al-rich garnets relative to Fe-rich garnets. In the alunitejarosite
minerals, 18O is enriched in the sulfate site in alunite relative to jarosite by approximately
2.4 ‰ at 250°C (Stoffregen et al. 1994, Rye and Stoffregen 1995).
EQUILIBRIUM FRACTIONATION FACTORS
Equilibrium isotopic fractionation factors are typically derived by one of three
methods: (1) experimental determination, (2) theoretical estimation using calculated bond
strengths or statistical mechanical calculations based on data on vibrational frequencies
of compounds, and (3) analysis of natural samples for which independent estimates of
temperature are available. Each method has advantages and disadvantages. Experimental
determination provides a direct measurement of the fractionation, but such efforts are
commonly hampered by experimental kinetic limitations and the fact that solutions used
in experiments often do not approximate natural solutions. Theoretical estimation avoids
the kinetic hurdles of experimental studies but is limited by the availability and accuracy
of data required for the estimation. Fractionation factors derived from the analysis of
natural materials provides a means of investigating isotopic fractionations when data
from neither of the other methods are available. However, this method is subject to
retrograde isotopic exchange and uncertainties related to the contemporaneity of the
mineral pairs and to the independent temperature estimate.
Available sulfur, oxygen, and hydrogen isotopic fractionation factors for sulfate
minerals are limited to a few mineral species, despite the geological importance of
numerous sulfates. Isotopic fractionation studies at temperatures ranging from
approximately 0 to 600°C have been conducted on dissolved and gaseous sulfate and
sulfite species, and for the minerals anhydrite, gypsum, barite, cha1canthite, mirabilite,
alunite, and jarosite. Oxygen and hydrogen isotope studies of sulfates have investigated
fractionation between crystallographically distinct sites as well. This chapter provides a
critical evaluation of the available fractionation data with the goal of assembling an
internally consistent set of sulfur, oxygen, and hydrogen isotope fractionation factors for
sulfate minerals.
Sulfur
Data for the sulfur isotope fractionation for sulfate minerals are limited. The gypsum-
dissolved sulfate system has been investigated by Thode and Monster (1965) and the
barite-dissolved sulfate system has been investigated by Kusakabe and Robinson (1977).
Both studies found limited sulfur isotope fractionation between the mineral and the
solution. At ambient temperature, Thode and Monster (1965) determined a sulfur isotope
fractionation between gypsum and dissolved sulfate of Δgypsum-sulfate = δ34Sgypsum ‒ δ34Ssulfate
= +1.65 ‰. At temperatures between 110 and 350°C, Kusakabe and Robinson (1977)
found no sulfur isotope fractionation between barite and dissolved sulfate (i.e., Δbarite-sulfate =
0 ‰). These results are consistent with the predictions of Sakai (1968) based on statistical
mechanics. Sakai (1968) concluded that because of the minimal modification of the
 Stable Isotope Systematics of Sulfate Minerals 551

vibrational energy of the sulfur atom in the SO42- radical by the crystalline field, the sulfur
isotope fractionation between a sulfate mineral and aqueous sulfate should be small. He
also concluded that a small, but measurable, enrichment in 34S in sulfate minerals should
follow CaSO4 > CaSO4·2H2O > SrSO4 > BaSO4 > PbSO4, which is in agreement with the
experimental data for gypsum by Thode and Monster (1965), and for barite by Kusakabe
and Robinson (1977). A major benefit of the minimal fractionation between sulfate
minerals and aqueous sulfate is that the measured isotopic composition of the sulfate
mineral approximates the isotopic composition of the parent fluid. This feature is important
in interpreting secular variations in the global oceanic sulfur cycle by using the sulfur
isotopic composition of marine evaporites as a record of ancient seawater sulfate
compositions.
From a process-oriented perspective, it is useful to be able to relate the isotopic
composition of sulfates to sulfides. Ohmoto and Lasaga (1982) critically evaluated the
available data on the fractionation of sulfur isotopes between SO42- and H2S in aqueous
systems (Robinson 1973, Bahr 1976, Sakai and Dickson 1978, Igumnov et al. 1977) to
eliminate suspect experimental data that may have experienced re-equilibration during
quenching. Ohmoto and Lasaga (1982) derived a temperature-dependent equation to
describe this sulfur isotope fractionation. Following the convention of Ohmoto and Rye
(1979), zero fractionation has been assumed between all sulfate minerals and aqueous
sulfate for the purpose of evaluating fractionation among sulfate minerals and H2S. Sulfur
isotope fractionations for aqueous sulfate and sulfate minerals relative to H2S are
presented in Table 1 and Figure 3. For reference, sulfur isotope fractionations for pyrite-
H2S and SO2-H2S from Ohmoto and Rye (1979) are also presented (Table 1; Fig. 3).
Oxygen
Data are available for oxygen isotope fractionation between the sulfate oxygen and
-
water for the compounds aqueous sulfate (SO42- and HSO4 ), anhydrite (CaSO4), barite
(BaSO4), alunite (KAl3(SO4)2(OH)6), and jarosite (KFe3(SO4)2(OH)6). Fractionation factors
between hydroxyl oxygen and water are available for alunite and jarosite, as are the
intramineral fractionation factors for SO42--OH-. Oxygen isotope fractionation factors
between crystallographic H2O and water are also available for gypsum (CaSO4-2H2O) and
mirabilite (Na2SO4·10H2O). To improve internal consistency among the oxygen isotope
fractionation factors presented in Table 1, published data for fractionations relative to H2O
were recalculated using a value of α CO 2 H 2 O = 1.04115 at 25°C (Brenninkmeijer et al. 1983)
where appropriate. In addition, data from studies that employed concentrated salt solutions

Figure 3. Temperature dependence

of equilibrium sulfur isotope fraction-
ation factors relative to H2S for
dissolved sulfate, sulfate minerals,
SO2, and pyrite. The dashed line
indicates a 0.0 ‰ 1000 1nα value.
Data from Table 1.

TABLE 1 HERE
552 Seal, Alpers & Rye

Table 1. Equilibrium isotopic fractionation factors for sulfate minerals

and related compounds.

Compound or A B C T(°C) Data

component range* sources
i(-j)
Sulfur
A × 10 B × 10
6 3

10001nα i − H S =
2 2
+ + C (T in K)
T T
Sulfate minerals 6.463 0.56 200 – 400 2
and aqueous
sulfate
SO2 4.70 -0.5 350-1050 1
FeS2 0.40 200 – 700 1

Oxygen
A × 10 B × 10
6 3

10001nα i − H O =
2 2
+ + C (T in K)
T T
-
SO42- (HSO4 ) 3.26 -5.81 70 – 350 5, 6, 7
Anhydrite 3.21 -4.72 100 – 550 8
Barite 2.65 -4.97 110 – 350 5, 9
Alunite SO4 3.09 -2.94 250-450 3
Jarosite SO4 3.53 -6.91 100 – 250 4
Alunite OH 2.28 -3.9 250 – 450 3
Jarosite OH 2.1 -8.77 100 – 250 4

Oxygen
A × 10 B × 10
6 3

10001nα i − j = 2
+ + C (T in K)
T T
Alunite SO4-OH 0.8 0.96 250 – 450 3
Jarosite SO4-OH 1.43 1.86 100 – 250 4

* Temperature range refers to the experimental temperature range; note that fractionation factors may extrapolate
significantly beyond these ranges (see text)
Data sources: 1. Ohmoto and Rye (1979); 2. Ohmoto and Lasaga (1982); 3. Stoffregen et al. (1994); 4. Rye and
Stoffregen (1995); 5. This study; 6. Lloyd (1968); 7. Mizutani and Rafter (1969a); 8. Chiba et al. (1981); 9. Kusakabe
and Robinson (1977).

were also corrected for isotopic effects of salt hydration using data from Horita et al.
(1993, 1995).
-
Oxygen isotope exchange between aqueous sulfate, both as SO42- and HSO4 , and
water has been investigated by Lloyd (1968) and Mizutani and Rafter (1969a) at
temperatures from 70 to 350°C (Fig. 4). Both studies determined similar fractionations
between aqueous sulfate and water. Thus, the expression presented in Table 1 and Figures
4, 5, and 6 represents a linear regression of both data sets. For anhydrite, mineral-water
oxygen isotope fractionations have been studied by Lloyd (1968) and Chiba et al. (1981)
with quite different results. The 1000 1nα factors for anhydrite-water of Lloyd (1968) are
between 2.5 and 7.0 ‰ greater than those of Chiba et al. (1981) for the temperature range
from 100 to 550°C (Fig. 5). Chiba et al. (1981) cited data from natural samples, which
 Stable Isotope Systematics of Sulfate Minerals 553

Figure 4. Comparison of the temperature

dependence of oxygen isotope
fractionation factors between dissolved
sulfate and water from Mizutani and
Rafter (1969a) and Lloyd (1968). The
solid line is the linear regression of both
data sets. The dashed line indicates a 0.0
‰ 1000 1nα value.

Figure 5. Comparison of the temperature

dependence of oxygen isotope fractionation
factors between simple sulfate minerals and
water. Anhydrite (L) is from Lloyd (1968);
Anhydrite (C) is from Chiba et al. (1981). Barite
is modified from Kusakabe and Robinson (1977).
SO4 curve is the best-fit curve from Figure 4 and
Table 1. The dashed line indicates a 0.0 ‰ 1000
1nα value.

support their experimental fractionation factors rather than those of Lloyd (1968). In
addition, the anhydrite-water fractionations of Chiba et al. (1981) are more consistent with
the theoretical fractionations of anhydrite relative to SO4 and barite (Sakai 1968) than those
of Lloyd (1968). Therefore, the fractionation factors of Chiba et al. (1981) are preferred
over those of Lloyd (1968). Kusakabe and Robinson (1977) presented experimentally
determined fractionations factors between barite and water. Their mineral-water
fractionations for barite relative to published values for anhydrite are consistent with what
would be expected on the basis of atomic mass and ionic radius differences between Ba2+
and Ca2+ (cf. O’Neil et al. 1969). The fractionation expression for the barite-water system
in Table 1 is derived from the experimental work of Kusakabe and Robinson (1977), with a
correction for the revised 25°C α CO 2 − H 2 O value and hydration effects.
Stoffregen et al. (1994) presented oxygen isotope fractionation factors for the systems
- -
alunite SO42--water, alunite OH -water, and intramineral alunite SO42--alunite OH (Table 1;
Fig. 6). Likewise, Rye and Stoffregen (1995) presented similar fractionation data for
jarosite (Table 1; Fig. 6). In both cases, explicit corrections for hydration effects were not
incorporated in the expressions for isotopic fractionations because the data are lacking to
554 Seal, Alpers & Rye

make these corrections. However, preliminary experiments by Stoffregen et al. (1994)

indicated that deviations from pure water due to salt effects in the alunite SO42--water
fractionation reach 4 ‰ at 400°C in 1.0 m KCl + 0.5 m H2SO4 solutions. It was also
determined that K/Na substitutions in alunite have negligible effect on SO42- oxygen
isotope fractionation in the temperature range 350 to 450°C.

Figure 6. Compilation of recommended equilibrium oxygen isotope

fractionation factors relative to H2O for various compounds, mineralogical sites,
and dissolved sulfate. See text for data sources.

Gonfiantini and Fontes (1963) determined that the oxygen isotope fractionation
between the hydrate water in gypsum and water is independent of temperature from 17 to
57°C and 1000 1nα gypsum − H 2 O ≈ Δ gypsum − H 2 O = 3.7 ‰ (Fig. 6). Stewart (1974) determined
that the oxygen isotope fractionation between hydrate water in mirabilite and solution
(1000 1nα mirabilite− H 2 O ≈ Δ mirabilite− H 2 O ) is 2.0 ‰ at 0°C and 1.4 ‰ at 25°C (Fig. 6). The
experimental solutions were saturated with respect to mirabilite. On the basis of the data
presented by Horita et al. (1993), the effect of the hydration of Na2SO4 on the oxygen
isotope fractionation between mirabilite and water should be less than the analytical
uncertainty; therefore, no correction for hydration effects is necessary to relate the
fractionation factors to pure H2O.
Hydrogen
Mineral-water hydrogen isotope fractionation factors have been experimentally
determined for alunite, jarosite, gypsum, mirabilite, and chalcanthite. Hydrogen isotope
fractionation factors generally tend not to be described by simple expressions following the
form of Equation (9). Therefore, hydrogen isotope fractionation factors are presented only
in Figure 7 without accompanying equations. Stoffregen et al. (1994) experimentally
measured hydrogen isotope fractionation factors between alunite and aqueous solutions and
found 1000 1nα alunite − H 2 O values that ranged from -19 at 450°C to -6 at 250°C. Bird et al.
(1989) proposed an alunite-water fractionation of 4 ‰ at ambient temperatures on the basis
of data from natural samples. Rye and Stoffregen (1995) experimentally measured
hydrogen isotope fractionation factors between jarosite and aqueous solutions and found
1000 1nα jarosite − H 2 O (≈ Δ jarosite − H 2 O ) values that were generally independent of temperature
 Stable Isotope Systematics of Sulfate Minerals 555

Figure 7. Equilibrium hydrogen isotope fractionation factors relative to H2O

for various minerals. The 25°C jarosite fractionation factor is from Alpers et
al. (1988). Other data sources cited in text.

from 150 to 250°C and averaged -50 ‰. Alpers et al. (1988) proposed a jarosite-water
fractionation of -59 ‰ at ambient temperature on the basis of data from a jarosite that
precipitated from mine water in the laboratory. Stoffregen et al. (1994) determined that
any variations in alunite-water fractionations associated with hydration effects of
dissolved salts were within the analytical uncertainty.
Fontes and Gonfiantini (1967) investigated hydrogen isotope fractionation between
gypsum and water in solutions saturated with gypsum; the 1000 1nα gypsum − H 2 O values were
independent of temperature from 17 to 57°C and averaged -15 ‰ (Fig. 7). Stewart (1974)
determined hydrogen isotope fractionations (10001nα mirabilite − H 2 O ) between mirabilite and
water at 0 and 25°C and found them to be 18.8 and 16.9 ‰, respectively (Fig. 7). For both
the gypsum and mirabilite studies, the compositions of the experimental solutions were not
sufficiently dilute that they would not be expected to affect mineral-water fractionations
due to hydration effects (Horita et al. 1993). Heinzinger (1969) measured hydrogen isotope
fractionations of 1000 1nα chalcanthite − H 2 O (≈ Δ chalcanthite − H 2 O ) = -22.7 ‰ between chalcanthite
and a saturated aqueous solution at ambient temperature. Data are lacking to evaluate the
effect of copper salt hydration on hydrogen isotope fractionation. In addition, Heinzinger
(1969) determined that there is a 57 ‰ intramineral fractionation in chalcanthite between
one hydrogen-bonded water molecule and the four water molecules bonded to Cu2+.
Geothermometry
Stable isotope geothermometry is typically based on the partitioning of stable isotopes
between two substances, such as sulfur between barite and pyrite. However, the
partitioning of an isotope between two crystallographically distinct sites within a single
mineral, such as oxygen between the sulfate and hydroxyl sites in alunite, can be used as a
single mineral geothermometer (Pickthorn and O’Neil 1985, Rye et al. 1992, Rye and
Alpers 1997). The use of stable isotopes for geothermometry is based on several
requirements or assumptions. First, the minerals must have formed contemporaneously in
equilibrium with one another. Second, subsequent re-equilibration or alteration of one or
both minerals must not have occurred. Third, pure minerals must be separated for isotopic
556 Seal, Alpers & Rye

analysis. Fourth, the temperature dependence of the fractionation factors must be known. In
addition, greater precision in the temperature estimate will be achieved from the use of
mineral pairs that have the greatest temperature dependence in their fractionations. Kinetic
considerations offer both advantages and disadvantages to geothermometry. Rapid kinetics
of isotope exchange promote mineral formation under equilibrium conditions.
Unfortunately, rapid exchange kinetics also make mineral pairs prone to re-equilibration
during cooling. In contrast, sluggish kinetics hamper isotopic equilibration between
minerals. However, once equilibrated, mineral pairs with sluggish exchange kinetics, such
as sulfate-sulfide mineral pairs will tend to record peak formation conditions without
subsequent re-equilibration at lower temperatures.
Coexisting pairs of sulfate minerals are uncommon and the temperature dependence
of sulfur and oxygen isotope fractionation are similar among sulfate minerals (Figs. 3 and
6). Thus, thermometry based on coexisting sulfate minerals is not practical.
Geothermometry based on isotopic data from sulfate minerals commonly employs sulfur
isotope fractionation between sulfate minerals (most commonly barite or anhydrite) and
sulfide minerals (most commonly pyrite), or oxygen isotope fractionation between sulfate
minerals and other oxygen-bearing minerals, such as silicates or carbonates. Intramineral
fractionation of oxygen isotopes between SO4 and OH in alunite allows single mineral
geothermometry. Oxygen isotope fractionation between dissolved SO42- and H2O, and
sulfur isotope fractionation between dissolved SO42- and H2S have been used to assess
reservoir temperatures in geothermal systems.
Ohmoto and Lasaga (1982) evaluated the kinetics of sulfur isotope exchange between
SO42- and H2S and found that exchange rates were dependent on pH, temperature, and total
concentrations of dissolved sulfur. Rates generally increase with increasing temperature
and sulfur concentration, and with decreasing pH. It was found that for “typical”
hydrothermal systems of near-neutral to slightly acidic conditions (pH ≈ 4 to 7) and ΣS =
10-2 m, isotopic equilibrium was unlikely to reached at temperatures below 200°C in
geologically reasonable time periods (Fig. 8). Therefore, sulfate-sulfide sulfur isotope
thermometry should be most useful in the study of acidic (acid-sulfate) and high-
temperature magmatic hydrothermal systems, and the study of the metamorphism of
sulfate-bearing rocks. An excellent example of acidic settings is in the case of alunite from

Figure 8. Comparison of the kinetics of sulfur (Ohmoto and Lasaga 1982)

and oxygen (Chiba and Sakai 1985) isotope exchange in terms of pH vs
logt1/2. Modified from Cole and Ohmoto (1986).
 Stable Isotope Systematics of Sulfate Minerals 557

acid-sulfate hydrothermal systems, for which the pH is sufficiently low to permit isotopic
equilibration at temperatures below 250°C, and alunite-pyrite pairs can be used for
exceptionally good geothermometry (Rye et al. 1992, Deen et al. 1994).
Because of the relationships expressed in Equations (4) and (5), mineral-mineral
fractionation equations can be derived from the equations in Table 1. An equation to
calculate the temperature recorded by a coexisting pair of barite (Ba) and pyrite (Py) can
be derived as follows:
1000 1nα Ba − Py ≈ Δ Ba − Py = δ34SBa − δ34SPy (10)
Thus,
Δ Ba − Py = Δ Ba − H 2S − Δ Py − H 2S = δ34SBa − δ34SH 2S − (δ34SPy − δ34SH 2S ) (11)
or
Δ Ba − Py ≈ 1000 1nα Ba − H 2S − 1000 1nα Py − H 2S (12)
Substituting from Table 1 yields
6.463 × 106 ⎛ 0.40 × 106 ⎞
Δ Ba − Py = + 0.56 ⎜⎜
− ⎟⎟ (13)
T2 ⎝ T 2
⎠
with T in K, or,
6.063 × 106
Δ Ba − Py = + 0.56 (14)
T2
Solving for T, and converting to °C yields:

6.063 × 106
T(°C ) = − 273.15 (15)
Δ Ba − Py − 0.56
For example, a sample with δ34SBa = 21.0 ‰ and δ34SPy = 5.1 ‰, a temperature of
356°C is calculated using Equation (15). Temperatures calculated using Equation (15) are
significantly lower (~70°C) than those calculated using the equations presented by
Ohmoto and Rye (1979). The difference is the result of the revised equation for SO42--
H2S sulfur isotope fractionation presented by Ohmoto and Lasaga (1982). Thus, data
from geothermometry studies published using the older fractionation factors should be
recalculated using the updated fractionation factors before reinterpretation. Figure 9
compares temperatures calculated on the basis of the revised SO42--H2S fractionation
factors of Ohmoto and Lasaga (1982) with those calculated using the earlier fractionation
factors summarized by Ohmoto and Rye (1979).

PROCESSES THAT CAUSE STABLE ISOTOPIC VARIATIONS

Variations in the stable isotopic composition of natural systems can result from a
variety of equilibrium and kinetically controlled processes, which span a continuum. The
most important steps for producing sulfur and oxygen isotopic variations in sulfate
minerals are the geochemical processes that initially produce (or transform) the sulfate
from (to) other sulfur species, rather than precipitation of the sulfate mineral from dissolved
sulfate. The more important processes generally involve the oxidation of sulfide to sulfate,
or the reduction of sulfate to sulfide. In addition, the low-temperature kinetic rates of many
of the oxidation and reduction processes are enhanced by bacterial mediation, which can
impart distinct isotopic fractionations to these processes. Thus, the complex aqueous
geochemistry of sulfur species is a key aspect of understanding the stable isotope
558 Seal, Alpers & Rye

Figure 9. Comparison of sulfate-pyrite sulfur isotope temperatures based on SO4‒H2S

fractionation factors of Ohmoto and Lasaga (1982) and those summarized by Ohmoto
and Rye (1979). X-axis is the temperature calculated using the SO4‒H2S fractionations
of Ohmoto and Lasaga (1980) (O&L). ΔT is the difference between the temperatures
calculated using the fractionation equations of Ohmoto and Lasaga (1980) and the low-
temperature (solid) and high-temperature (dashed) fractionation equations summarized
in Ohmoto and Rye (1979) (O&R) from the literature.

geochemistry of sulfate minerals. Ohmoto (1972) developed the principles for application
of sulfur isotope systematics to hydrothermal ore deposits. Van Stempvoort and Krouse
(1994) provided a comprehensive review of controls on oxygen isotope compositions in
aqueous sulfate. Taylor and Wheeler (1994) reviewed controls of the oxygen and sulfur
isotope systematics of acid mine-drainage systems. Comprehensive reviews of the controls
on the oxygen and sulfur isotope systematics of sulfates in ore deposits have been given by
Ohmoto and Rye (1979), Ohmoto (1986), and Ohmoto and Goldhaber (1997).
If the starting reservoir is limited in size, many equilibrium and kinetic processes can
be described as Rayleigh distillation processes. Rayleigh processes are described by the
equation
R = Ro ƒ(α-1) (16)
where Ro is the initial isotopic ratio, and R is the isotopic ratio when a fraction (ƒ) of the
starting amount remains, and α is the fractionation factor, either equilibrium or kinetic.
This equation can be recast in the δ notation for sulfur isotopes as
δ34S = (δ34So + 1000) ƒ(α-l) ‒1000 (17)
Rayleigh models accurately describe isotopic variations associated with processes such
as the precipitation of minerals from solutions, the condensation of precipitation (rain,
snow) from atmospheric moisture, and the bacterial reduction of seawater sulfate to sulfide,
among others. Bacterial reduction of seawater sulfate can be modeled using Equation (17).
If α=1.0408 and δ34So =21.0 ‰, then precipitation of pyrite from H2S produced from
bacterial reduction of sulfate will preferentially remove 32S and the first pyrite formed will
have δ34S ≈ -20 ‰. The preferential removal of 32S will cause the δ34S of the residual
aqueous sulfate to increase, which in turn will lead to an increase in the δ34S of
 Stable Isotope Systematics of Sulfate Minerals 559

Figure 10. Rayleigh distillation curves for

bacterial reduction of seawater sulfate
showing the change in δ34S of resultant H2S
(filled circles), residual sulfate (open
circles), and bulk sulfide (X) as a function of
reaction progress. Pyrite precipitated from
the H2S would be expected to have a δ34S
that is approximately 4 ‰ higher than that
shown for H2S assuming equilibrium
fractionation between pyrite and H2S.
Modified from Ohmoto and Goldhaber
(1997). All isotopic values in permil
(VCDT).

subsequently formed pyrite (Fig. 10). Under closed-system behavior, after all of the
sulfate has been reduced, the bulk isotopic δ34S of the precipitated pyrite will equal the
δ34S of the initial sulfate. However, the δ34S of individual pyrite grains or growth zones
can be both lower and higher than the bulk composition, depending on when they formed.
Mixing is another important process that can cause isotopic variations and it can be
modeled on the basis of simple mass-balance equations such as
δmixture = XAδA + XBδB (18)
where δmixture is the resulting isotopic composition of the mixture, δA and δB are the
isotopic compositions of components A and B, and XA and XB are the mole fractions of
components A and B.
Kinetics of isotope exchange reactions
Ohmoto and Lasaga (1982) found that exchange rates between dissolve SO42- and H2S
decreased proportionally with increasing pH at pH < 3; from pH ≈ 4 to 7, the rates remain
fairly constant; and at pH > 7, the rate also decreases proportionally with increasing pH.
Ohmoto and Lasaga (1982) proposed that the overall rate of exchange is limited by
exchange reactions involving intermediate valence thiosulfate species (S2O32-), the
abundance of which is dependent on pH. The rate-limiting step was postulated to be an
intramolecular exchange between non-equivalent sulfur sites in thiosulfate. They calculated
the most rapid equilibration rates at high temperature (T = 350°C) and low pH (pH ≈ 2) of
approximately 4 hours for 90 percent equilibrium between aqueous sulfate and sulfide;
however, at low temperature (T = 25°C) and near neutral pH (pH = 4-7), the time to attain
90 percent equilibrium reached 9 × 109 years. Disequilibrium between sulfate and sulfide
minerals is prevalent in hydrothermal and geothermal systems below 350°C (Figs. 8, 11).
The rate of oxygen isotope exchange between dissolved sulfate and water has been
investigated by Hoering and Kennedy (1957) from 10 to 100°C, by Lloyd (1968) from 25
to 448°C, and by Chiba and Sakai (1985) from 100 to 300°C. Both Lloyd (1968) and
Chiba and Sakai (1985) used Na2SO4 solutions in which pH was adjusted using either
H2SO4 or Na(OH). The pH values of the experimental solutions of Lloyd (1968) ranged
from 3.8 to 9.0, and those of Chiba and Sakai (1985) ranged from 2.3 to 7.3. In contrast,
560 Seal, Alpers & Rye

Figure 11. Comparison between observed Δ and

equilibrium Δ between SO4 and H2S as a function
of T for mineral deposits and geothermal systems.
White rectangles are from porphyry hydrothermal
systems, black rectangles are from seafloor
massive sulfide hydrothermal systems, gray
rectangles are from Mississippi Valley-type
hydrothermal systems, and white circles are from
modem geothermal systems. Equilibrium curve is
from Ohmoto and Lasaga (1980). Diagram is
modified from Ohmoto and Lasaga (1980). All
isotopic values in permil (VCDT).

the experimental solutions of Haering and Kennedy (1957) consisted of mixtures of

H2SO4 and H2O that ranged from 1.04 to 16.0 M H2SO4 ; although the resulting pH was
not specified, it can be modeled using the computer program PHRQPITZ for electrolyte
solutions of high ionic strength (Plummer et al. 1988). Calculated pH values range from
0.0 to less than -25. Pitzer interaction parameters for H2SO4 solutions have been
calibrated up to 6 M and appear to be valid up to approximately 9 M (Ptacek and Blowes,
this volume). Thus, the present study is limited to a discussion of data from Hoering and
Kennedy (1957) for concentrations less than 9 M (pH > -5) even though the higher
concentration data follow the same general trends as the lower concentration data in term
of rate of exchange versus pH. Comparison of the three experimental studies in terms of
rate of reaction versus pH at various temperatures shows excellent agreement between the
studies of Hoering and Kennedy (1957) and Chiba and Sakai (1985) at 100°C, the only
temperature of overlap between the two studies (Fig. 12). In contrast, the study of Lloyd
(1968) covers the entire temperature range of both of the other studies, but the rates of
exchange derived from Lloyd (1968) are approximately two orders of magnitude faster
than the rates of Hoering and Kennedy (1957) and Chiba and Sakai (1985). Insufficient
details are provided in the three studies to determine which studies are most likely to be
correct; however, the agreement between the results of Hoering and Kennedy (1957) and
Chiba and Sakai (1985) is noteworthy.
Chiba and Sakai (1985) concluded that exchange is limited by interactions between
H2SO4° and H2O at low pH- (pH = 2.3 to 2.7 at T = 100°C and pH = 4.0 to 5.5 at T =
200°C), and between HSO4 and H2O at higher pH (pH = 6.1 to7.3 at T = 300°C). It was
also concluded that the rate of sulfate-water oxygen isotope exchange is sufficient to
permit this pair to be used as geothermometer for geothermal systems. In contrast,
sulfate-water isotopic exchange in ambient seawater (pH ≈ 8; T ≈ 4°C) was suggested to
require on the order of 109 years to reach equilibrium. At very low pH, corresponding to
extreme acid mine-drainage conditions, tentative extrapolation of the data of Chiba and
Sakai (1985) suggested that a 50 % approach to equilibrium could be achieved in a period
of several months (Taylor and Wheeler 1994).
A comparison of exchange rates of sulfur isotopes between sulfate and sulfide, and
oxygen isotopes between sulfate and water indicates that the half-life of exchange for
oxygen is several orders of magnitude faster than for sulfur at acidic to neutral conditions
 Stable Isotope Systematics of Sulfate Minerals 561

Figure 12. Comparison of the kinetics of oxygen isotope exchange data

between dissolved sulfate and water in terms of pH and logtl/2 (hours). Data
from Hoering and Kennedy (1957), Chiba and Sakai (1985), and Lloyd (1968).
Values of pH for solutions of Hoering and Kennedy (1957) were computed
using PHRQPITZ (Plummer et al. 1988) with MacInnes convention for scaling
individual ion activity coefficients. Note agreement between Hoering and
Kennedy (1957) and Chiba and Sakai (1985) studies at 100°C.

(Fig. 8). This difference is the result of differences in rate-limiting steps. In sulfate, sulfur
atoms are surrounded by tetrahedrally coordinated oxygen atoms. Thus, multiple O-S
bonds must be broken before sulfur isotope exchange can occur, whereas only a single O-
S bond must be broken for oxygen isotope exchange. In addition, Ohmoto and Lasaga
(1982) concluded that the rate-limiting step for sulfur isotope exchange is the
intramolecular exchange between non-equivalent sites in thiosulfate.
Sulfate reduction, sulfide oxidation, and associated processes
The processes of sulfate reduction and sulfide oxidation in natural systems tend to
have associated characteristic, kinetically controlled, non-equilibrium oxygen and sulfur
isotope fractionations. Oxidation and reduction processes can occur in both biotic and
abiotic environments. Isotopic variations associated with the biogenic reduction of sulfate
have been studied by numerous researchers, most of whom have concentrated on the role
of the dissimilatory sulfate-reducing bacteria such as Desulfovibrio desulfuricans. The
activity of sulfate-reducing bacteria in marine sediments throughout most of geological
time had a profound effect of the sulfur isotope composition of seawater sulfate, which is
discussed in a later section.
Sulfate-reducing bacteria are active only in anoxic environments such as below the
sediment-water interface, and in anoxic water bodies. Various species of sulfate-reducing
bacteria can survive over a range of temperature (0 to 110°C) and pH (5 to 9.5)
conditions, but prefer near-neutral conditions and can withstand a range of salinities from
dilute up to halite saturation (Postgate 1984). The metabolism of sulfate-reducing bacteria
can be described by the general reaction:
-
2 CH2O + SO42- = H2S + 2 HCO3 (19)
where CH2O represents generic organic matter (Berner 1985). The H2S can be lost to the
water column, reoxidized, fixed as pyrite, or other sulfide minerals if reactive metals are
present, or fixed as organic-bound sulfur. In near surface sediments deposited under
normal (oxygenated) marine settings, the activity of sulfate-reducing bacteria is limited
562 Seal, Alpers & Rye

by the supply and reactivity of organic matter; in freshwater and euxinic basins, the
activity is limited by sulfate availability (Berner 1985).
The fractionation of sulfur isotopes between sulfate and sulfide during bacterial sulfate
reduction is a kinetically controlled process in which 34S is enriched in the sulfate relative
to the sulfide, in the same sense as equilibrium fractionation between sulfate and sulfide
(Chambers and Trudinger 1979). The sulfate-reducing bacteria more readily metabolize 32S
and 16O relative to 34S and 18O, respectively. Thus, the δ34S and δ18O of the residual
aqueous sulfate increase during the reaction progress. The fractionation associated with
bacterial sulfate reduction (1000 1nαSO4 −H2S ) typically ranges from 15 to 60 ‰ (Goldhaber
and Kaplan 1975) in marine settings, compared to an equilibrium, abiotic fractionation of
approximately 73 ‰ at 25°C. The magnitude of the fractionation has been shown to be a
function of the rate of sulfate reduction, which can be related to sedimentation rates. The
smaller fractionations (~15‰) correspond to faster rates of sulfate reduction and
sedimentation, whereas the larger fractionations (~60‰) correspond to slower rates of
sulfate reduction and sedimentation (Goldhaber and Kaplan 1975; Fig. 13).

Figure 13. Observed sulfur isotope fractionation (Δpyrite-SO4) associated with bacterial
sulfate reduction as a function of rate of reduction and rate of sedimentation. Modified
from Ohmoto and Goldhaber (1997). Abbreviations: BS, Black Sea; C, Carmen Basin;
EC, East Cortez Basin; LI, Long Island Sound; MR, Marina Del Rey Harbor; P,
Pascadero Basin; RS, Red Sea; SB, Santa Barbara Basin; SL, Solar Lake; SG, Spencer
Gulf. Curved line represents best fit to the data. All isotopic values in permil.

The fractionation of oxygen isotopes during bacterial reduction of sulfate is due to a

complex combination of processes among which are preferential reduction of isotopically
light enzyme (adenosine 5′-phosphosulfate; APS)-bound sulfate to sulfite (SO32-),
reoxidation of sulfite to APS and sulfate, sulfite-water oxygen isotope exchange, and
reoxidation of sulfide (Van Stempvoort and Krouse 1994). Experimental studies to
 Stable Isotope Systematics of Sulfate Minerals 563

quantify the oxygen isotope fractionation associated with bacterial sulfate reduction have
been conducted by Mizutani and Rafter (1969b, 1973), and Fritz et al. (1989). Whereas the
sulfur isotope systematics associated with bacterial sulfate reduction follow a Rayleigh
pathway, the oxygen isotope systematics do not (Fritz et al. 1989). The kinetics of direct
oxygen isotopic exchange, as discussed above, are sufficiently slow at ambient, near-
neutral conditions to exclude direct sulfate-water exchange as a significant process during
bacterial sulfate reduction. However, the exchange of oxygen isotopes between sulfite and
water at ambient conditions (pH < 9) is extremely rapid, on the order of minutes (Betts and
Voss 1970, Holt et al. 1981). Oxygen isotope exchange between sulfite and water, and the
subsequent reoxidation of the sulfite to sulfate, seem to be important processes in
determining the δ18O of residual sulfate during bacterial reduction (Mizutani and Rafter
1969b, Fritz et al. 1989). Experimental studies using 35S-labeled H2S demonstrate the
viability of concurrent reduction of SO42- and oxidation of H2S in anoxic environments
(Trudinger and Chambers 1973, Elsgaard and Jørgensen 1992). The net result is that even
though sulfate-reducing bacteria preferentially metabolize 16O relative to 18O to produce a
kinetic fractionation ranging between 25 and 29 ‰ (depending on the rate of reduction),
the resulting δ18O of the residual sulfate will reflect the δ18O of the water if there is
sufficient time for oxygen isotope exchange to occur between the aqueous sulfite and water
(Mizutani and Rafter 1969b, 1973; Fritz et al. 1989).
In a study of the abiotic (thernl0chemical) reduction of aqueous sulfate through high-
temperature (200 to 350°C) interactions with Fe2+ (fayalite and magnetite), Shanks et al.
(1981) found that the sulfur isotope data could be modeled as an equilibrium Rayleigh
distillation process. The δ34S of residual aqueous sulfate increased during reduction in
accordance with published equilibrium fractionation factors (Ohmoto and Lasaga 1982;
Table 1). Similarly, Sakai et al. (1980) found that sulfur isotope fractionations associated
with thermochemical reduction of dissolved sulfate through reaction with olivine (XFo =
0.90) at 400°C produced results consistent with equilibrium exchange between sulfate
and sulfide. Orr (1974) and Kiyosu (1980) documented sulfur isotopic effects associated
with thermochemical reduction of sulfate due to interaction with organic matter and
found that sulfate-sulfide kinetic fractionation was less than 10 ‰. No studies have been
conducted to examine the oxygen isotope systematics associated with abiotic reduction of
sulfate. However, on the basis of the rate of oxygen isotopic exchange between SO42- and
H2O at elevated temperature, sulfate-water oxygen isotopic equilibrium would be
expected at temperatures above 250°C (Chiba and Sakai 1985).
The supergene oxidation of sulfide minerals to form sulfate minerals or aqueous
sulfate produces distinct effects in terms of sulfur and oxygen isotope signatures. With
respect to sulfur isotopes, low-temperature oxidative alteration of sulfide minerals to sulfate
minerals is a quantitative, unidirectional process that seems to produce negligible sulfur
isotope fractionation; the δ34S of resulting sulfate minerals is indistinguishable from that of
the parent sulfide mineral (Gavelin et al. 1960, Nakai and Jensen 1964, Field 1966, Rye et
al. 1992). Gavelin et al. (1960) and Field (1966) documented similar sulfur isotope
composition among hypogene ore-sulfide minerals and various associated supergene
sulfate minerals, including gypsum, alunite, jarosite, anglesite, and brochantite. Field
(1966) proposed that the similarity of δ34S of sulfate minerals formed in the supergene
environment with that of the hypogene sulfide minerals from which they formed could be
used to distinguish supergene sulfates from high-temperature, hypogene sulfates, which
tend to form in isotopic equilibrium with the associated sulfides. A similar conclusion can
be reached regarding the relationship of aqueous sulfate with sulfide minerals in acid mine-
drainage settings. Taylor and Wheeler (1994) and Seal and Wandless (1997) found no
discernible difference between δ34S in the parent sulfides and the associated dissolved
sulfate. However, some experimental and field studies, particularly those involving sulfide
564 Seal, Alpers & Rye

oxidation in neutral to alkaline conditions have documented negative sulfate-sulfide

fractionations up to approximately 14 ‰ (Toran and Harris 1989), indicative of a kinetic
isotope fractionation. Another notable exception occurs when oxygenated mine waters
subsequently become exposed to anoxic conditions, and bacterial sulfate reduction occurs.
In these settings, the δ34S of dissolved sulfate is higher than that of the parent sulfide
minerals (Hamlin and Alpers 1996, Seal and Wandless 1997).
In contrast, the δ18O of mineral or aqueous sulfate formed during the biotic or abiotic
oxidative weathering of sulfide minerals depends on the source of oxygen. The greatest
attention has been focused on the weathering of pyrite because of its common occurrence
in numerous rock types. The weathering of pyrite is generally described by two
unidirectional reactions that differ in their oxidizing agent (O2 versus Fe3+) and their
source of oxygen (O2 and H2O). The two reactions are (Garrels and Thompson 1960,
Singer and Stumm 1970, Nordstrom and Alpers 1999a):
2−
FeS2 + 7/2 O 2(aq) + H 2O → Fe 2+ + 2 SO 4 + 2 H + (20)
and
2−
FeS2 + 14 Fe3+ + 8 H 2O → 15 Fe 2+ + 2 SO 4 + 16 H + (21)
The reaction rate of the second reaction is limited by the rate of the reaction
Fe 2+ + 1/4 O 2(aq) + H + → Fe3+ + 1/2 H 2O (22)
which is greatly enhanced by the presence of the iron-oxidizing bacteria, such as
Thiobacillus ferrooxidans (Singer and Stumm 1970). Other bacteria, such as Thiobacillus
thiooxidans, which oxidize sulfur, can also accelerate the rate of pyrite oxidation. Even
though Equations (20) to (22) effectively describe the mass balance of pyrite oxidation,
they do not define the pathways of electron transfer (Toran and Harris 1989). Equations
(20) and (21) only describe the overall of pyrite weathering; actual rate-limiting steps
may involve a variety of intermediate species including native sulfur (S°), sulfite (SO32-),
thiosulfate (S2O32-), and polythionates (SnO62-; Moses et al. 1987). The stable isotope
systematics accompanying these reactions have been most recently reviewed by Taylor
and Wheeler (1994) and Van Stempvoort and Krouse (1994).
Because of the sluggish kinetics of sulfate-water oxygen isotope exchange, aqueous
sulfate should preserve, except under the most acidic conditions, a record of the sources of
oxygen from which the oxygen originated. For the first reaction (Eqn. 20), 87.5 % of the
oxygen is derived from O2 and 12.5 % is derived from H2O on a molar basis, whereas in
the case of the second reaction (Eqn. 21), 100 % of the oxygen is derived from water. The
δ18O of atmospheric oxygen is globally homogeneous and averages 23.5±0.3 ‰
(Kroopnick and Craig 1972), whereas the δ18O of water varies considerably depending on
whether it is meteoric water or seawater, or if it has been modified by processes such as
evaporation. In addition, kinetic fractionations are associated with the oxidation of pyrite
and the incorporation of oxygen from O2 or H2O into sulfate. Taylor et al. (1984)
conducted an experimental study of oxygen isotope fractionation associated with pyrite
oxidation under a variety of conditions and proposed that the relationship between the δ18O
of dissolved sulfate and water could be used to constrain contributions of atmospheric-
derived and water-derived oxygen to the weathering of pyrite. Taylor and Wheeler (1994)
presented the following equation to describe the δ18O of sulfate produced from the
oxidation of pyrite in terms of relative contributions of the reactions portrayed in Equations
(20) and (21):
δ18OSO4 = X (δ18Ow + εw) + (1 − X)[0.875(δ18OO2 + εO2 ) + 0.125(δ18Ow + εw)] (23)
where X is the proportion of oxidation accomplished through Reaction (21), 1-X is the
 Stable Isotope Systematics of Sulfate Minerals 565

proportion of oxidation accomplished through Reaction (20), and εw and ε O2 are the
isotopic enrichment factors for the incorporation of oxygen from water and atmospheric
oxygen, respectively. Because of the presence of H2O in Reaction (20), the permissible
range of water-derived oxygen is between 12.5 and 100 mole % (Fig. 14). Taylor and
Wheeler (1994) emphasized that the use of Equation (23) assumes that the oxygen isotopic
enrichment factors are constant and that isotope exchange between aqueous sulfate and
water does not occur after the oxidation of the pyrite. This model also assumes that the
participation of O2 and H2O is wholly by molecular pathways. However, oxygen may also
be involved in an abiotic or biotic electrochemical pathway involving only electron transfer
(cf. Toran and Harris 1989). Nevertheless, interpretation of the oxygen isotope systematics
of sulfate derived from the oxidation of pyrite using Equation (23) can provide useful
insights into the environment of pyrite weathering as has been attempted for supergene
oxidation in some ore deposits (Rye et al. 1992, Rye and Alpers 1997).

Figure 14. Oxygen isotope variation

between sulfate and water associated with
the oxidation of pyrite. Modified from
Taylor and Wheeler (1994). Dashed
isopleths mark mixtures between Reactions
(20) (0%) and (21) (100%) as described by
Equation (23). Fields for mine drainage
sites are indicated by the abbreviations:
L Leviathan;
A Argo Tunnel;
WS West Shasta district;
SM mines in Great Smoky
Mountains National Park;
CB Cabin Branch mine;
S Sulfur mine; and
PM Penn mine.
Data from Taylor and Wheeler (1994), Seal and Wandless (1997), and Hamlin and Alpers (1996). Unshaded
fields indicate sites where evaporation was significant, whereas shaded fields indicate sites where evaporation
was insignificant. Vectors show the general trends expected for evaporation and oxygen isotopic exchange
between SO4 and H2O. All isotopic values in permil.

Taylor et al. (1994) determined oxygen isotopic enrichment factors associated with
the oxidation of pyrite to form sulfate under a variety of conditions. The origins of these
fractionations are complex and involve a variety of processes including mediation by
microbes, isotopic exchange between water and intermediate sulfur oxyanions, and
isotopic effects associated with the microenvironments of pyrite oxidation, among others.
Lloyd (1967) experimentally investigated the oxidation of aqueous Na2S and H2S. Under
acidic conditions, Taylor et al. (1984) found that ε O2 varied between experiments
conducted under sterile conditions (-4.3 ‰) and those conducted in the presence of
Thiobacillus ferrooxidans (-11.4 ‰). The latter enrichment factor is more relevant to
natural systems. Experimental studies by Taylor et al. (1984) found εw to be 4.0 ‰.
Comparison with data from natural systems led Taylor and Wheeler (1994) to propose
that εw = 0 ‰ may be more appropriate. For neutral to alkaline environments, ε O2 = -8.7
‰ seems to be more appropriate (Lloyd 1967, Taylor and Wheeler 1994). Greater
complexity to the stable isotope systematics associated with sulfide oxidation occurs at
higher pH values, at which exchange reactions with intermediate sulfur oxyanions can be
important (Goldhaber 1983). Under acidic conditions, the rate of pyrite oxidation is
relatively rapid and the reaction tends to be complete at the pyrite surface (Wiersma and
Rimstidt 1984, McKibben and Barnes 1986).
566 Seal, Alpers & Rye

Because of the considerable overlap in the magnitude of sulfate-water fractionations

found in experimental studies, Δ SO4 −H2O values cannot be used to quantify the relative
importance of biotic and abiotic pyrite oxidation (Toran and Harris 1989, Taylor and
Wheeler 1994). However, higher Δ SO4 −H2O values correlate qualitatively with biotic, aerobic
environments that cycle between wet and dry conditions, whereas lower Δ SO4 −H2O values
correlate with sterile, anaerobic, submersed conditions (Taylor and Wheeler 1994).
Mechanisms of precipitation and dissolution of sulfate minerals
The precipitation mechanism for sulfate minerals can have important implications for
the interpretation of stable isotope data for these minerals (Ohmoto and Goldhaber 1997).
At temperatures above 250°C, dissolved sulfate and sulfide can be expected to be in sulfur
isotope equilibrium under most circumstances. Rapid cooling, followed by precipitation of
sulfate minerals may not be recorded in sulfate isotope systematics because the dissolved
species may fail to re-equilibrate, thereby preserving disequilibrium Δ SO4 −H2S values that are
too low for the precipitation conditions (Ohmoto and Lasaga 1982, Rye 1993). Simple
cooling of hydrothermal fluids is not a likely mechanism for the precipitation of sulfate
minerals because of their retrograde solubility. However, below approximately 75°C,
barite, anglesite, and celestite show prograde solubility (Rimstidt 1997). Coprecipitation of
sulfide and sulfate minerals occurs typically when decreasing temperature is accompanied
by an increase in sulfate concentration at the site of deposition. Local increases in sulfate
can result from the oxidation of H2S, sulfide minerals, or native sulfur, the hydrolysis of
SO2, or mixing with a sulfate-rich fluid (Ohmoto and Goldhaber 1997). The rapid rates of
these processes may preclude the re-establishment of isotopic equilibrium. Thus, isotopic
disequilibrium can be the result of processes at the immediate site of sulfate-mineral
deposition.
The hydrolysis, or disproportionation of SO2 is a related process that can be important
for coprecipitation of sulfide and sulfate minerals. Above 400°C, SO2 is the dominant
oxidized species of sulfur in hydrothermal fluids (Holland 1965, Burnham and Ohmoto
-
1980). During cooling, the hydrolysis of SO2 produces both H2S and HSO4 through the
reaction:
-
4 H 2O + 4 SO 2 ↔ H 2S + 3 H + + 3 HSO 4 (24)
The relationships of these processes to specific geochemical settings are discussed below.
The dissolution of sulfate minerals generally proceeds as a quantitative disequilibrium
process. Thus, dissolution produces no fractionation of neither sulfur or oxygen isotopes.
The retrograde solubility of many sulfate minerals tends to cause dissolution during cooling
when associated sulfide or silicate minerals are actively precipitating (Rimstidt 1997).
GEOCHEMICAL ENVIRONMENTS
Secular variations in seawater sulfate isotopic compositions
The geochemistry of sulfur has been intimately tied to global-scale factors such the
oxidation state of the atmosphere, anoxia state of the oceans, and the rate of erosion on the
continents, which are ultimately influenced by tectonic processes. The sulfur and oxygen
isotope signatures of marine evaporitic sulfate minerals provide a record of secular
variations in the oceans. Claypool et al. (1980) summarized published sulfur and oxygen
isotopic data from marine evaporite deposits, and contributed new data from other key
parts of the geological record. Since 1980, several studies have concentrated on Proterozoic
and Archean isotopic variations (Hayes et al. 1992,Ohmoto 1992, Strauss 1993, Ross et al.
1995). Paytan et al. (1998) recently reported a high-resolution record for Cenozoic seawater
sulfate from deep-sea sediments. Mineralogically, these data are derived dominantly from
 Stable Isotope Systematics of Sulfate Minerals 567

gypsum, anhydrite, and barite. Collectively, these studies and the studies referenced therein
have found that the mean δ34S of marine evaporitic sulfate has varied from around 4 ‰ at
3.4Ga to a high of 33 ‰ during Cambrian time to a low of about 10 ‰ during Permian and
Triassic time, ultimately to a modern value around 23 ‰ (Fig. 15). The oxygen isotopic
data reported in the literature are more limited in scope. The mean δ18O of marine
evaporitic sulfate has varied from around 17 ‰ at 900 Ma to a low of 10 ‰ during
Permian time, to a modern value of 13 ‰ (Fig. 16). On the basis of known low-
temperature fractionation factors between gypsum and water, the δ34S of the associated
seawater sulfate should be 1.65 ‰ less than that of the mineral (Thode and Monster 1965),
and the δ18O of the associated dissolved sulfate should 3.7 ‰ less than that of the mineral
(Gonfiantini and Fontes 1963). The δ34S of modern seawater is homogeneous on a global
scale, with δ34S = 20.99±0.08 ‰ (Rees et al. 1978). The homogeneity results from the
rapid mixing time of seawater (~1000 y) relative to the residence time of sulfate in
seawater (8 Ma; Holland 1978). The δ18O of modern seawater sulfate is also homogeneous

Figure 15 (left). Secular variations of the δ34S marine evaporites. Curves were compiled from
Claypool et al. (1980), Strauss (1993), and Paytan et al. (1998). Curve shows best fit to the data.
Shaded field shows range of most likely values for the Phanerozoic and part of the Proterozoic.
Filled circles from 3.3 to 3.5 Ga are from marine evaporites; horizontal bars from 1.9 to 2.5 Ga are
from trace sulfate minerals in marine carbonates and cherts (Strauss 1993). All isotopic values in
permil (VCDT).
Figure 16 (right). Secular variations of the δ18O of marine evaporites. Modified from Claypool
et al. (1980). Black curve shows best fit to the data; shaded field shows range of values. All
isotopic values in permil (VSMOW).
568 Seal, Alpers & Rye

throughout the oceans with a value of 9.5 ‰ (Longinelli and Craig 1967, Nriagu et al.
1991).
Claypool et al. (1980) and Holser et al. (1979) interpreted qualitatively the δ34S and
18
δ O of marine sulfate in terms of a dynamic balance between (1) the removal of sulfate
from seawater either by its reduction to sulfide by sulfate-reducing bacteria, or its
precipitation as sulfate minerals in marine evaporite deposits and (2) the addition to
seawater of sulfate derived either from the oxidative weathering of sulfide minerals on
the continents, or from the weathering of evaporite minerals on the continents (Fig. 17a).
Oxygen isotopic exchange between dissolved sulfate and water is not considered to be a
complicating factor because of the slow rate of exchange at neutral to alkaline conditions
relative to the residence time of sulfate in the oceans (Hoering and Kennedy 1957, Chiba
and Sakai 1985). Holser et al. (1979) also concluded that the rapid oxidation-reduction
cycling mechanism proposed by Lloyd (1968), and the hydrothermal circulation of
seawater at mid-ocean ridges were not significant processes affecting seawater-sulfate
stable isotope systematics.

Figure 17. Covariation of oxygen and sulfur isotopes in global oceans. (a) Vector
model for processes contributing to secular variations in stable isotopes for marine
evaporites. Modified from Claypool et al. (1980); (b) Secular covariations of marine
evaporitic δ34S and δ18O. Modified from Claypool et al. (1980). Age along the curve is
indicated in Ma. All isotopic values in permil (VSMOW or VCDT).

The oceanic sulfur-budget analysis of Claypool et al. (1980) made several assump-
tions about sulfur and oxygen isotope fractionations associated with the various input and
output processes, some of which may require modification in light of other studies. It was
 Stable Isotope Systematics of Sulfate Minerals 569

assumed that the sulfate-sulfide sulfur isotope fractionation associated with bacterial
sulfate reduction was 40 ‰. The sulfate-sulfide fractionation in modern sediments
averages 40 ‰, but has been documented to range between 15 and 60 ‰ depending on
the rate of sulfate reduction, which can be related to sedimentation rate (Goldhaber and
Kaplan 1975). It was also assumed that the weathering of pyrite and other sulfide
minerals generally occurs without significant sulfur isotope fractionation, which seems to
be supported by more recent studies (e.g. Taylor and Wheeler 1994). Claypool et al.
(1980) estimated that the modern mean δ34S of sulfate input derived from the weathering
of pyrite is -17±3 ‰. Likewise, they assumed that the removal of sulfate as sulfate
minerals causes a decrease of 1.65 ‰ in the δ34S of the residual aqueous sulfate and that
the generation of dissolved sulfate derived from the weathering sulfate minerals produces
no significant fractionation of sulfur isotopes. It was estimated that the modern mean δ34S
of sulfate input derived from the weathering of evaporites is 16±2 ‰.
Holser et al. (1979) assumed on the basis of the experimental study of Mizutani and
Rafter (1969b) that the effect on the residual dissolved sulfate of oxygen isotopic
fractionation associated with bacterial sulfate reduction was approximately 25 % of that
for sulfur (i.e. 10 ‰). Mizutani and Rafter (1969b) found that the ratio δ18O of variations
to δ34S variations with the progress of sulfate reduction was (δ18O-δ18Oi)/( δ34S-δ34Si) =
1/3.8, where δ18Oi and δ34Si are the starting oxygen and sulfur isotopic compositions,
respectively, of the dissolved sulfate, and δ18O and δ34S are the compositions at the
termination of the sulfate reduction experiments. In contrast, data from the experimental
studies of Mizutani and Rafter (1973) and Fritz et al. (1989) yielded (δ18O-δ18Oi)/( δ34S-
δ34Si) ratios of 1/2.5 and 1/19.4, respectively; all three studies combined yield a ratio of
1/6.1 (Fig. 18). The significant differences among these three studies in terms of

Figure 18. Covariation δ34S and δ18O in residual sulfate associated with
bacterial sulfate reduction in laboratory (Mizutani and Rafter 1969a,
Mizutani and Rafter 1973, Fritz et al. 1989) and field (Zak et al. 1980)
studies. Data sets are normalized to their starting sulfur (δ34Si) and oxygen
(δ18Oi) compositions. Lines are least-squared fits to the data. Dashed line
indicates δ18O-δ18Oi=0.0 ‰. All isotopic values in permil (VSMOW or
VCDT).
570 Seal, Alpers & Rye

temperature, culture media, bacterial cultures, and experiment duration make it difficult to
determine which data might be most applicable to natural systems. However, Zak et al.
(1980) reported δ18O and δ34S data for pore waters from modern marine sediments in the
Pacific Ocean which yield a ratio of 1/2.1 associated with sulfate reduction in natural
settings. This ratio suggests that a fractionation of 19 ‰ may be more appropriate to
describe oxygen isotope variations associated with bacterial sulfate reduction in the oceans.
The hydrothermal barite of the Creede district, Colorado, indicates that aqueous
sulfate in the saline waters of ancient Lake Creede underwent an O/S isotope enrichment
during reduction of about 1/2 prior to recharge into the hydrothermal system, but the δ18O
values may have been affected by exchange with highly evaporated (18O-enriched) waters
(Rye et al. 1988, Barton et al. 2000). The covariation of sulfur and oxygen isotopes
during bacterial sulfate reduction warrants further research.
For inputs, Holser et al. (1979) assumed that the oxidative weathering of pyrite
introduces sulfate to the oceans with an average δ18O of 2.0 ‰ on the basis of
experimental studies on the oxidation of aqueous sulfides. Subsequent studies on the
oxidation of pyrite, summarized by Taylor and Wheeler (1994), indicate that most
weathering of pyrite occurs through processes dominated by 50 to 75 % of Reaction (21),
which derives all of its oxygen from water. This range corresponds to a resultant sulfate
δ18O between -1 and 3 ‰, assuming global mean precipitation δ18O of -4.0 ‰ (Craig and
Gordon 1965). Claypool et al. (1980) assumed that the weathering of evaporites results in
no fractionation of oxygen isotopes in sulfates, and assumed that the global mean δ18O of
evaporitic sulfate was 12 ‰. They proposed that the combined input of sulfate to the
oceans from the weathering of pyrite and evaporites has a δ18O of 5 ‰.
Claypool et al. (1980) concluded that the secular variations for the marine sulfate
curves for δ34S and δ18O share some similarities, but also differ significantly. With the
exception of Permian time, the secular variations are most dominant in the δ34S record with
a range of over 20 ‰ (excluding the Permian), whereas the δ18O record has a range of 5 ‰
or less (excluding the Permian) (Fig. 17b). These coupled variations were interpreted in
terms of variable inputs and output to the oceans. The large excursions in δ34S, with
subordinate variations in δ18O, require that episodes of erosion and oxidation of pyrite be
accompanied by erosion and dissolution of evaporites, and episodes dominated by sulfate
reduction be accompanied by precipitation of evaporitic sulfate (Fig. 17a,b). To date, the
role of hydrothermal circulation through ridge crests on the sulfur and oxygen isotopic
systematics of seawater has yet to be addressed comprehensively.
The secular variations of marine sulfate and sedimentary sulfide minerals have been
central to discussions of the evolution of the early atmosphere. The bulk δ34S of Earth is
assumed to be similar to meteoritic δ34S (i.e. 0 ‰). The δ34S of the earliest marine sulfide
minerals clusters around 0 ‰ (prior to 2.7 Ga) and is generally interpreted to reflect the
absence of significant concentrations of dissolved sulfate due to the absence of an
oxygenated atmosphere (Lambert and Donnelly 1990, Hayes et al. 1992). Both factors
would have caused significant scatter in the isotopic composition of marine sulfides
because of the large fractionations between oxidized and reduced sulfur species. On the
basis of sulfur isotope data from marine sulfate and sulfide minerals, Ohmoto and Felder
(1987) proposed a sulfate-sulfide kinetic fractionation of approximately 10 ‰ for bacterial
sulfate reduction in the Archean oceans. Ohmoto (1992) suggested that the smaller kinetic
fractionation was due to a higher rate of sulfate reduction, which can be related to inferred
higher ocean temperatures (30 to 50°C) and greater availability of oxidizable organic
matter. A smaller kinetic fractionation would necessitate a greater degree of sulfate
reduction to produce larger variations in the δ34S of marine sulfide minerals. However, the
presence of sulfate in Archean oceans does not necessarily require the presence of free
 Stable Isotope Systematics of Sulfate Minerals 571

oxygen in the atmosphere.

Seafloor hydrothermal systems
Modern systems. The stable isotope
characteristics of modern seafloor hydrothermal
systems from mid-ocean ridges have been
summarized recently by Shanks et al. (1995).
Studies which present stable isotope data for
sulfate minerals for individual seafloor
hydrothermal systems include those of Shanks and
Seyfried (1987), Zierenberg and Shanks (1988,
1994), Alt et al. (1989), Peter and Shanks (1992),
Goodfellow and Franklin (1993), and Zierenberg et
al. (1993). Among the minerals studied were
anhydrite, gypsum, and barite. Most studies have
concentrated on the δ34S of seafloor sulfate
minerals with limited attention given to δ18O. All
of these studies are from mid-ocean ridge
hydrothermal systems, with the exception of that of
Zierenberg and Shanks (1988), which described the
Red Sea hydrothermal system, for which Miocene
evaporite sequences may have been an important
component of the sulfate budget.
In mid-ocean ridge hydrothermal systems,
sulfate is stripped from seawater during heating
associated with downwelling due to the retrograde
solubility of anhydrite and other sulfate minerals
(Bischoff and Seyfried 1978, Seyfried and
Bischoff 1981, Shanks et al. 1995). Thus, sulfate
minerals at the vent site form from the interaction
of H2S-bearing hydrothermal fluids with SO4-
bearing ambient seawater. The clustering of the
δ34S of anhydrite from off-axis basalts in the
Pacific at around 21 ‰, the composition of
modern seawater (Fig. 19), is consistent with the Figure 19. δ34S histogram for sulfide and
nearly quantitative stripping of sulfate during sulfates minerals from modern seafloor
recharge of hydrothermal convection systems (Alt hydrothermal systems. Shaded vertical
field shows the isotopic composition of
et al. 1989). In general, the δ34S values of modern modern seawater. Vectors show the effect
mid-ocean ridge seafloor vent-site sulfate minerals of sulfate reduction and oxidation of H2S
cluster around the δ34S of modern seawater, but on the δ34S of sulfate exposed to these
locally range from a low of 3.7 ‰ to a high of processes. Data are from Shanks and
29.3 ‰ (Fig. 19). Janecky and Shanks (1988) Seyfried (1987), Zierenberg and Shanks
have modeled chemical and sulfur isotopic (1988), Peter and Shanks (1992), and
Goodfellow and Franklin (1993). Data
processes in seafloor hydrothermal systems. It was for DSDP 504B are for trace sulfide and
found that models based on sulfate-sulfide sulfate minerals in altered oceanic crust
disequilibrium best describe the sulfur isotope (Alt et al. 1989). All isotopic values in
systematics of modern seafloor hydrothermal permil (VCDT).
systems, and that barite and anhydrite form under
distinctly different thermal regimes. Both minerals
form from the mixing of vent fluids with ambient
572 Seal, Alpers & Rye

seawater. Barite forms during conductive cooling with mixing, whereas anhydrite forms
either during adiabatic mixing or during conductive heating with mixing. Scatter from
seawater values for both minerals fails to correlate with these inferred differences in
depositional environments. Thus, sulfate minerals with a δ34S less than seawater are
generally interpreted to reflect the oxidation of hydrothermal H2S (typical range δ34S = -3
to 8 ‰; Shanks et al. 1995) in the vent complex. Sulfate minerals with a δ34S greater than
that of seawater are generally interpreted to reflect the residual sulfate from
thermochemical reduction through interactions with ferrous silicates or organic matter in
the shallower portions of the hydrothermal system (Fig. 19).

Figure 20. Sulfur and oxygen isotope data for sulfate minerals and dissolved sulfate
from the Red Sea. Red Sea brine and Miocene evaporite data are from Zierenberg
and Shanks (1986); epigenetic vein and brine-pool precipitate data are from
Zierenberg and Shanks (1988). Dashed line represents proposed mixing of sulfate
between the upper brine and a hypothetical hydrothermal fluid to produce the lower
brine. The line for the hypothetical hydrothermal sulfate was calculated using the
fractionation factors in Table 1 and assuming equilibrium with seawater (δ18O = 1.2
‰) and H2S (δ34S = 5.4 ‰) at various temperatures. The gray field represents
hypothetical mixing between the lower brine and dissolved sulfate derived from
Miocene evaporites. Note that many of the epigenetic veins and brine-pool
precipitates fall outside of the mixing field. The bold arrows represent the trend
expected for sulfate reduction. All isotopic values in permil (VSMOW and VCDT).

Studies from seafloor hydrothermal systems in the Red Sea illustrate the complex
interplay of geochemical processes that control the stable isotope compositions of sulfate
minerals. Zierenberg and Shanks (1986) observed distinct differences in the δ34S and δ18O
of dissolved sulfate among the upper and lower brines of the Atlantis II deep, which hosts
impressive modern massive sulfide deposits, and the brines of the nearby Valdivia and
Suakin deeps. The variations were interpreted in terms of a combination of hightemperature
oxygen isotope exchange between seawater and dissolved sulfate, and mixing of seawater
sulfate with sulfate derived locally from Miocene evaporite deposits; the dissolved sulfate
of the brines from the Valdivia and Suakin deeps are dominated by simple mixtures of
seawater sulfate and Miocene evaporite sulfate, without high-temperature isotope
exchange. In contrast, the brines of the Atlantis II deep show significant stratification with
the lower brine having a lower δ18O and a slightly higher δ34S compared to those of the
upper brine. The δ18O and δ34S of dissolved sulfate from the upper brine are
indistinguishable from modern seawater (Fig. 20). Therefore, the sulfate from the lower
 Stable Isotope Systematics of Sulfate Minerals 573

brine probably comprises a mixture of Red Sea deep water and hydrothermal sulfate. The
isotopic composition of the hydrothermal end-member sulfate can be constrained by
calculating the isotopic composition of aqueous sulfate in equilibrium with local seawater
(δ18O = 1.2 ‰; Zierenberg and Shanks 1986) and hydrothermal H2S (δ34S = 5.4 ‰;
Zierenberg and Shanks 1988) at various temperatures (Fig. 20). A mixing line drawn from
the sulfate of the upper brine (δ34S = 21.0 ‰, δ18O = 9.5 ‰) through that of the lower brine
(δ34S = 21.4 ‰, δ18O = 7.5 ‰) intersects the hypothetical end-member sulfate curve at δ34S
= 22.2 ‰, δ18O = 3.6 ‰, and T = 360°C. Thus, on the basis of mass-balance calculations
(Eqn. 18), approximately 35 % of the sulfate in the lower brine may have been derived
from the hydrothermal fluid. The inferred temperature is consistent with temperatures
commonly found in modern vent systems (Shanks et al. 1995) and with mineralogical and
fluid inclusion evidence for a high temperature history for the Atlantis II deep
hydrothermal fluids (Pottorf and Barnes 1983, Oudin et al. 1984). The assumption of
equilibrium is reasonable with respect to the kinetics of isotopic exchange of SO4 with H2O
and H2S at these temperatures. However, less certain is the ability of the inferred
hydrothermal fluids to transport significant amounts of dissolved sulfate because of the
retrograde solubility of anhydrite.
Regardless of the ultimate origin of the sulfate in the lower brine, the oxygen and
sulfur isotopic composition of anhydrite and barite in epigenetic vein and brine-pool
precipitates in the Atlantis II deep are only partly explained by mixing of sulfate from the
lower brine pool and sulfate from Miocene evaporites. The δ18O of the lower brine and the
evaporites encloses the range of the sulfate minerals from the epigenetic veins and brine-
pool precipitates, but the δ34S of the minerals ranges to higher δ34S values than for a simple
mixture. One possible explanation for the higher δ34S values is that the sulfate had
undergone partial thermochemical reduction prior to deposition as anhydrite or barite.
Thermochemical reduction of sulfate produces an increase in the δ34S of the residual sulfate
(Shanks et al. 1981). The corresponding change in δ18O associated with thermochemical
reduction has not been investigated in natural or experimental systems, as discussed above,
but can be expected to result in an increase in the δ18O in the residual sulfate, consistent
with bacterial sulfate reduction. If a δ34S/δ18O slope of 4/1 is chosen to represent the trend
produced by sulfate reduction, then the partial reduction of various admixtures of lower
brine sulfate and evaporite sulfate can adequately describe the isotopic variations of the
anhydrite and barite in the epigenetic veins and brine-pool precipitates. No bacterial
reduction of sulfate is known to occur at present in the Atlantis II deep (Zierenberg and
Shanks 1988), but low δ34S compositions in oxide-rich facies of metalliferous sediments
indicate such reduction in the past (Shanks and Bischoff 1980). The average δ34S of sulfide
minerals from the Atlantis II deep (~5.4 ‰; Shanks et al. 1980) is significantly higher than
the δ34S of basalts such as the ones underlying the Red Sea. Thus, the elevated δ34S of
sulfide minerals indicates a component of thermochemical sulfate reduction.
Ancient systems. Ancient seafloor hydrothermal systems produce two general classes
of deposits containing sulfate minerals: volcanic-associated (volcanogenic) massive sulfide
deposits, and sedimentary-exhalative (sedex) massive sulfide deposits, which Maynard and
Okita (1991) refer to as “cratonic rift-type barite deposits”. Volcanic-
associated deposits form at active mid-ocean ridge spreading centers, and in arc-volcanic
rocks, continental rifts, and Archean greenstone belts, whereas sedex deposits form in
failed continental rift settings. Maynard and Okita (1991) further distinguish “continental
margin type barite deposits”, which form in convergent margins and typically are lacking
associated Zn and Pb. Volcanic-associated sedex deposits can have associated
hydrothermal sulfate minerals, usually anhydrite, barite, or gypsum. Gypsum is typically a
weathering product of primary anhydrite. In general, barite and anhydrite occur as veins,
disseminations in massive sulfide ore, or as massive stratiform zones. Sulfur isotope data
574 Seal, Alpers & Rye

are reasonably common, but oxygen

isotope results are more limited.
The stable isotope systematics of
volcanic-associated massive sulfide
deposits and their significance have been
reviewed by Ohmoto (1986) and Huston
(1999). The sulfur-isotope data for sulfate
minerals plotted in Figure 21 are from
published studies of deposits that range in
age from the Recent back to
approximately 1.9 Ga. This figure
includes data from the Kuroko district,
Japan (Watanabe and Sakai 1983,
Kusakabe and Chiba 1983), the Eastern
Black Sea district, Turkey (Ҫagatay and
Eastoe 1995), the Buchans camp,
Newfoundland (Kowalik et al. 1981),
various districts in California (Eastoe and
Gustin 1996), the Mount Read volcanic
belt, Tasmania (Solomon et al. 1969,
1988; McGoldrick and Large 1992,
Gemmell and Large 1992), Barite Hill,
South Carolina (Seal et al., in press), the
Skellefte district, Sweden (Rickard et al.
1979), and Buttle Lake, British Columbia
(Seccombe et al. 1990). For the most part,
the δ34S of sulfate minerals from these
deposits is similar to the δ34S of coeval
seawater, although deviations to higher
and lower values can be found (Fig. 21).
Some of the deviations from the coeval
seawater composition may be attributed
to uncertainties in the age of the deposits,
particularly for those deposits that formed Figure 21. Histograms of the difference between the
near abrupt transitions in the secular sulfur isotope composition of sulfate mineral and that of
curve for marine sulfate isotopic coeval seawater from ancient seafloor hydrothermal
compositions (Fig. 15). The similarity deposits. The δ34S of seawater was taken from the average
indicates that sulfate minerals associated values of Claypool et al. (1980) and adjusted assuming a
with seafloor hydrothermal systems are mineral-dissolved sulfate fractionation of 1.65 ‰. See text
for sources of data. Huston (1999) produced a similar
generally produced by the mixing of diagram to compare sulfide and sulfate mineral
sulfate-rich ambient seawater with Ba- or compositions relative to seawater. All isotopic values in
Ca-bearing hydrothermal fluids. As with permil (VCT).
modern hydrothermal systems, values
lower than that of ambient seawater are typically interpreted to represent the oxidation of
hydrothermal H2S to produce sulfate for later precipitation, such as at Buttle Lake, British
Columbia (Seccombe et al. 1990).
Values higher than that of ambient seawater are typically interpreted to reflect partial
reduction of seawater sulfate and associated isotopic enrichment of the residual sulfate,
either by bacterial or abiotic processes, such as in the Mount Read volcanic belt, Tasmania
(Solomon et al. 1969, 1988; McGoldrick and Large 1992, Gemmell and Large 1992). The
δ34S of sulfate minerals from the deposits of the Mount Read volcanic belt is up to 22 ‰
 Stable Isotope Systematics of Sulfate Minerals 575

greater than that of coeval seawater, with a distinct mode around 14 ‰ greater than coeval
seawater (Fig. 21). Such pervasive reduction of seawater sulfate is unlikely to occur
through abiotic, thermochemical interactions with ferrous iron during hydrothermal
convection as suggested by Green et al. (1981) and Solomon et al. (1988), because of the
limited ability of hydrothermal fluids to transport seawater sulfate due to the retrograde
solubility of anhydrite and other sulfate minerals (Bischoff and Seyfried 1978, Seyfried and
Bischoff 1981). The common appearance of isotopic equilibrium between sulfate and
sulfide minerals likely reflects the dynamic, disequilibrium mixing environment between
hydrothermal fluids and ambient seawater typically invoked to explain the formation of
seafloor exhalative barite bodies. Bacterial reduction of seawater sulfate in anoxic bottom
waters is an alternative to explain the high δ34S of the Mount Read sulfates relative to
coeval seawater. Goodfellow and Jonasson (1984) proposed a similar mechanism to
explain enriched isotopic compositions of barite in the Selwyn Basin, Yukon relative to
coeval seawater. The deposits of the Mount Read volcanic belt formed at a time of oceanic
bottom-water anoxia in the Iapetus Ocean, which may have been of global extent (Leggett
1980). The presence of black slates associated with the Rosebery deposit in the Mount
Read belt supports this hypothesis, but the presence of minor hematite in some of the facies
of mineralization suggests that further evaluation may be warranted (Green et al. 1981).
The Selwyn Basin, Yukon, Canada hosts several syngenetic sedimentary exhalative
Zn-Pb-barite deposits and bedded barite deposits, which range in age from Cambrian to
Mississippian and formed from seafloor hydrothermal processes (Goodfellow and
Jonasson 1984). Goodfellow and Jonasson (1984) documented secular trends in the δ34S
of pyrite and barite within the basin. The Selwyn Basin barite compositions followed the
secular trends of global marine evaporites (Claypool et al. 1980), but were consistently
10 to 15 ‰ higher in δ34S. Goodfellow and Jonasson (1984) interpreted the higher δ34S of
the Selwyn barite relative to global evaporites to reflect formation in a restricted basin,
with anoxic bottom waters and limited access to the open ocean. The bacterial reduction
of sulfate caused the residual sulfate, now preserved as hydrothermal and sedimentary
barites, to become in δ34S relative to global averages.
“Continental margin type” bedded barite deposits, such as those in Nevada (Rye et al.
1978), in the Qinling district, China (Maynard and Okita 1991, Wang and Li 1991), and
Arkansas (Maynard and Okita 1991) tend to have δ34S values similar to that of coeval
seawater. In Nevada, in addition to bedded barite, barite also occurs in diagenetic rosette
and concretion morphologies. The δ34S of the rosettes is significantly higher than that of
coeval seawater (up to 30 ‰ greater) and the δ18O of sulfate shows a 1/3 enrichment slope
relative to δ34S (R.O. Rye unpubl. data). The high δ34S of these non-bedded barites was
interpreted to reflect the influence of closed-system bacterial sulfate reduction on residual
sulfate during diagenesis (Rye et al. 1978). Similarly, Nuelle and Shelton (1986) deter-
mined that nodular barites from Devonian shales in the Appalachian Basin, with δ34S up to
28 ‰ and δ18O up to15 ‰, were higher than coeval marine evaporites.
Magmatic systems
In the spring of 1982, a series of three major pyroclastic eruptions occurred at El
Chichón Volcano in eastern Mexico, involving an unusually sulfur-rich trachyandesitic
magma (Luhr et al. 1984, Rye et al. 1984). The sulfur-rich character was reflected in both
in the presence of microphenocrysts of anhydrite in fresh pumices and in the large size of
the stratospheric cloud of sulfur acid aerosols produced by the eruptions. Anhydrite has
since been recognized as a characteristic and important mineralogical component of fresh
tephra from Mount Pinatubo and other arc volcanoes that erupt oxidized, hydrous sulfurrich
magma (Fournelle 1990, Bernard et al. 1991, McKibben et al. 1996). Anhydrite is not well-
preserved in older pyroclastic rocks because it weathers out of tephra soon after eruption.
576 Seal, Alpers & Rye

Experimental studies have demonstrated the relatively high solubility of anhydrite and the
stability of dissolved SO42- species in oxidized melts (Carroll and Rutherford 1987, 1988,
Luhr 1990). It is therefore now commonly assumed that the anhydrite in these recent
pyroclastic materials is a juvenile magmatic phase. This was first demonstrated by a
combination of textural evidence and highly concordant temperatures based on sulfur and
oxygen isotopic compositions of anhydrite, pyrrhotite, magnetite and plagioclase crystals
from fresh pumices of the 1982 eruptions of El Chichón volcano which indicated a pre-
eruption temperature of 810±40°C. The isotopic composition of sulfate leached from fresh
ash-fall samples showed it to have been derived from a mixture of anhydrite micro-
phenocrysts and adsorbed sulfate, the latter derived from the oxidation of SO2 in the
eruption plume. The isotopic data also indicate the lack of early oxidation of significant
amounts of H2S, which should have dominated the eruption cloud based on H2S/SO2 in
equilibrium with the melt. Apparently, this oxidation occurs slowly as some sulfate aerosol
clouds have been observed to grow as they circled the Earth in the stratosphere (Bluth et al.
1997). The sulfur-rich magma may have developed from the assimilation of a volcanogenic
massive sulfide or sedimentary sulfate body. Finally, sulfur isotope data on sulfides and
anhydrite in lithic fragments support the development of a porphyry-type system overlying
the magma at El Chichón.
SHRIMP sulfur isotope studies of anhydrite in pumice from the early vent-clearing
June 1991 eruption of Mount Pinatubo, Philippines, show a bimodal sulfur isotope
distribution that may represent primary phenocrysts and xenocrysts derived from
hydrothermal anhydrite deposits occurring within the flanks of the volcano. To explain the
sulfur isotope systematics, McKibben et al. (1996) favored a mechanism involving long-
term passive degassing of sulfur from underlying basalt into the much more oxidized dacite
magma, and steady-state degassing of vapor enriched in 34S.
Igneous apatites can contain up to 1,500 mg/kg sulfur probably as SO42- substituting
for PO42-, and coexisting minerals such as biotite and magnetite may contain up to 200
mg/kg sulfur, probably as SH substituting for OH (Banks 1982). Various sulfate and sulfide
mineral inclusions may occur in minerals such as plagioclase and magnetite. Ueda and
Sakai (1984) used the different reactivity of minerals to various acids to recover separately
sulfate and coexisting sulfide in I-type igneous rocks. Their data seldom show sulfur
isotopic equilibrium. Lack of isotopic equilibrium between sulfate in apatite and sulfide in
igneous rocks has also been shown for a series of igneous rocks that culminated with the
intrusion of an anhydrite-bearing dike that interrupted ore deposition at Julcani, Peru (Deen
1990). This lack of equilibrium can be explained by a combination of assimilation of
country-rock coupled with fractionations due to degassing of magma during its ascent (See
also Ohmoto 1986).
Continental hydrothermal systems
Modern systems. The stable isotope systematics of active geothermal systems are
dominated by water, dissolved sulfate, and dissolved sulfide, depending on redox
conditions, and the interactions of these species with the reservoir rocks. The stable isotope
geochemistry of geothermal systems has been reviewed by Truesdell and Hulston (1980)
and Giggenbach et al. (1983a). Numerous studies have investigated the stable isotope
composition of dissolved sulfate in geothermal systems, particularly for oxygen relative to
H2O, and to a lesser extent, for sulfur relative to H2S. In addition, Rye et al. (1992)
presented data from alunite from active geothermal systems. Information recorded by
dissolved sulfate and dissolved sulfide depends on the temperature, pH, and on the
residence time of fluids in the system. Oxygen isotopes exchange more rapidly between
sulfate and water than do sulfur isotopes between sulfate and sulfide (Fig. 8). Kusakabe
(1974) and Robinson (1978) compared SO4-H2O oxygen isotope and SO4-H2S sulfur
 Stable Isotope Systematics of Sulfate Minerals 577

isotope geothermometers for geothermal systems in New Zealand and found that the
systematics of the oxygen isotopes were easier to interpret than those of the sulfur isotopes.

Figure 22. Δ versus 1000/T (K) for geothermal systems for oxygen isotopes between sulfate and
water, and for sulfur isotopes between sulfate and sulfide. Equilibrium curves were calculated from
the expressions in Table 1. “T reservoir” corresponds to temperatures estimated from geochemical
thermometers; “T surface” corresponds to surface temperatures of hot springs or well heads. See text
for sources of data. All isotopic values in permil. (VSMOW).

Figure 22 shows the relationship of SO4-H2O oxygen isotope fractionation and SO4-
H2S sulfur isotope fractionation to observed temperatures for geothermal systems in Japan
(Mizutani 1972, Sakai 1977, Sakai and Matsubaya 1977, Matsuo et al. 1985), Italy
(Cortecci 1974), New Zealand (Kusakabe 1974, Simmons et al. 1994), India (Giggenbach
et al. 1983b), United States (McKenzie and Truesdell 1977), and Mexico (Martinez et al.
1996). Temperatures are either surface temperatures of hot springs (<100°C), or reservoir
temperatures, determined by chemical thermometers (>100°C). Collectively, the SO4-H2O
oxygen isotope data appear to reflect reservoir temperatures that average ~ 200±75°C. The
good agreement among chemical thermometry (reservoir) temperatures and SO4-H2O
oxygen isotopic temperatures suggests the attainment of isotopic equilibrium, which forms
the basis of SO4-H2O oxygen isotope thermometry in geothermal systems (Cortecci 1974,
Sakai 1977, McKenzie and Truesdell 1977). Thus, the δ18O of dissolved sulfate in
geothermal systems is dominated by the δ18O of the water. The surface temperatures plot
below the equilibrium SO4-H2O oxygen isotope fractionation curve, and suggest that the
SO4-H2O pairs preserve a record of deeper and hotter reservoir conditions.
The δ34S values of SO4-H2S pairs are less consistent, with none of the data
approaching equilibrium for reservoir temperatures determined by various chemical
geothermometers (Fig. 22). For the Wairakei geothermal system, New Zealand, the
temperatures may reflect deep reservoir temperatures around 350°C (Kusakabe 1974). In
other systems, the inferred temperatures are unrealistically high (>450°C) for deep
reservoir temperatures. The interpretation of SO4-H2S sulfur isotopic data is also
complicated by shallow-level, disequilibrium oxidation of H2S, or reduction of sulfate, both
occurring with insufficient time to re-establish equilibrium (Kusakabe 1974, Robinson
1978, Sakai 1983).
Ancient systems. The stable isotope geochemistry of ancient continental hydrothermal
systems in igneous terranes is similar to that of modern geothermal systems, with the
578 Seal, Alpers & Rye

added, complicating factor that ancient systems typically represent the deeper, unexposed
portions of modern geothermal systems. Thus, in associated high-temperature regimes, SO2
(SO32-) dominates over SO3 (SO42-), and direct magmatic contributions of sulfur species
and water to the hydrothermal system are important. Low-temperature disequilibrium
processes dominate the stable isotope geochemistry of ancient continental hydrothermal
systems in sedimentary environments.
Many of the important geochemical processes that affect the stable isotope
characteristics of sulfate minerals in igneous hydrothermal systems are illustrated by the
minerals alunite [KAl3(SO4)2(OH)6] and jarosite [KFe3(SO4)2(OH)6] which occur in a
variety of mineral-deposit types. Alunite and jarosite are especially informative because
insights can be gained from four stable isotope sites by the analysis of δ34S from the SO4
site, δ18O from the SO4 and OH sites, and δD from the OH site (Rye et al. 1992, Rye and
Alpers 1997). In addition, because alunite and jarosite are K-bearing minerals, they are
useful as geochronometers (K/Ar, 40Ar/39AR; see Stoffregen et al., this volume). The
same processes are also recorded in simple sulfate minerals, such as barite and anhydrite,
in porphyry and polymetallic vein deposits. However, a key difference between acid-
sulfate environments that form alunite, and porphyry and polymetallic vein environments,
is that the highly acidic conditions associated with the formation of alunite are more
conducive to isotopic equilibrium among sulfur and oxygen species than the less acidic
conditions associated with the other deposit types.
Alunite can be found in four distinct settings known as the supergene, steam-heated,
magmatic hydrothermal, and magmatic steam environments (Rye et al. 1992). In contrast,
jarosite has only been found in steam-heated and supergene environments (Rye and
Alpers 1997); jarosite also occurs in the “sour gas” environment, in which sulfate
originates by oxidation of H2S, for example, in the Rio Grande rift in New Mexico and
Chihuahua, Mexico (Lueth et al. 2000). Other occurrences of hydrothermal jarosite are
summarized by Dutrizac and Jambor (this volume). Each environment displays unique
stable isotope characteristics and jarosite formation requires unique pH and ƒO2
conditions for formation (Stoffregen 1993).
For both supergene alunite and jarosite, the sulfur and oxygen isotope characteristics
of the sulfate are determined by the geochemical processes responsible for the weathering
of sulfide minerals, such as pyrite, as described in Reactions (20) and (21), and typically
reflect disequilibrium fractionations. For hydroxyl, the oxygen and hydrogen isotope
characteristics are dictated by the isotopic composition of the associated water, with the
principal difference between the two minerals being the -55±5 ‰ hydrogen isotope
fractionation between jarosite and water. Intramineral fractionation between the SO4 and
OH also typically reflects disequilibrium. Rye et al. (1992) and Rye and Alpers (1997)
defined a “supergene alunite SO4 field” (SASF), a “supergene jarosite SO4 field” (SJSF), a
“supergene alunite OH zone” (SAOZ), and a “supergene jarosite OR zone” (SJOZ) to
describe the isotopic compositions (Fig. 23) that would be expected for supergene alunites

Figure 23 (opposite page). Isotopic variations of alunite for various environments for given fluid
compositions. Modified from Rye et al. (1992). SASF indicates the Supergene Alunite Sulfate Field and
SAOZ indicates the Supergene Alunite OH Zone. Jarosites display similar isotopic variations (Rye and Alpers
1997), the principal difference being the large negative hydrogen isotope fractionation between jarosite and
water results in a much diminished supergene jarosite sulfate field (SJSF) which is the counterpart to the
SASF. The meteoric water line (MWL) and the field for hypothetical primary magmatic water (PMW) are
shown for reference. (a) oxygen and sulfur isotopic variations expected for supergene alunites; (b) hydrogen
and oxygen isotopic variations expected for supergene alunites; (c) oxygen and sulfur isotopic variations
expected for steam-heated alunites; (d) hydrogen and oxygen isotopic variations expected for steam-heated
alunites. Examples illustrate effects expected for an unmodified meteoric water (δD = -80 ‰, δ18O = -11.3 ‰)
and a modified meteoric water (δD = -110 ‰, δ18O = -5.0 ‰); (e) oxygen and sulfur isotopic variations
 Stable Isotope Systematics of Sulfate Minerals 579
expected for magmatic-hydrothermal alunites; (f) hydrogen and oxygen isotopic variations expected for
magmatic-hydrothermal alunites. The mixing trends shows the range isotopic effects expected for mixing
magmatic fluids with modified meteoric waters; (g) oxygen and sulfur isotopic variations expected for
magmatic-steam alunites; (h) hydrogen and oxygen isotopic variations expected for magmatic-steam alunites;
all isotopic values in permil (VSMOW or VCDT).
580 Seal, Alpers & Rye

and jarosites. The limits of supergene SO4 fields are defined by whether the oxygen used
to produce the sulfate was dominantly derived from water (Reaction 21) or atmospheric
oxygen (Reaction 20). The limits of the supergene OH zones are defined by temperature-
dependent mineral-water fractionations between geologically reasonable temperatures of
20 to 80°C. The sulfate isotope systematics of supergene alunite and jarosite can be
further complicated if the sulfate undergoes partial bacterial reduction prior to
precipitation. In this case, the resulting sulfate will have a higher δ18O and δ34S than its
starting composition as discussed above. Examples of supergene alunite and jarosite are
common throughout the world, and have been studied in Australia, Spain, New Mexico,
Arizona, Nevada, Colorado, Chile, and elsewhere; isotope compositions parallel the
meteoric water line over a wide range of latitudes (Rye et al. 1992, Rye and Alpers
1997).
Steam-heated environments share many geochemical characteristics with supergene
environments. Aqueous sulfate is derived from the oxidation, at or above the water table,
of H2S that is distilled off of hydrothermal fluids as described by the summary reaction:
H2S + 2 O2 = H2SO4 (25)
The reaction occurs at temperatures less than 100°C, but the resulting acids may
descend to greater depths at higher temperatures (Rye et al. 1992). The higher temperatures
and acidic conditions typical of geothermal fields promote isotopic equilibrium between
SO42- and H2O and unlike the supergene environment, the Δ18OSO4 −OH values for alunite
commonly give reasonable temperatures. Although sulfur isotopic equilibrium between
aqueous SO42- and H2S is not obtained in this environment, significant partial exchange
occurred before deposition of alunite at the Crofoot-Lewis deposit, Nevada (Ebert and Rye
1997). In geothermal and hydrothermal systems, the hydrothermal fluids can exchange
oxygen isotopes with country rocks to a significant degree, producing fluid with δ18O
values that are substantially higher than those of local meteoric waters. Thus, the predicted
isotopic fields for steam-heated alunite and jarosite can overlap those for supergene
minerals depending on the δ18O of the associated hydrothermal fluid (Fig. 23). Steam-
heated alunites and jarosites can generally be distinguished from supergene minerals by
their tendency for equilibrium fractionation between intramineral SO4-OH sites, and
inferred SO4-H2O pairs. Examples of steamed-heated alunites and jarosites can be found in
Nevada, Utah, Italy, New Zealand, and elsewhere (Rye et al. 1992, Rye and Alpers 1997,
Ebert and Rye 1997). Because of the exceptionally low pH and high ƒO2 required for
jarosite formation, steam-heated jarosites represent the last hydrothermal event in an area
and can be used to date the duration of hydrothermal mineralization or to recognize a later
hydrothermal event (Rye and Alpers 1997).
Magmatic hydrothermal alunite forms from sulfuric acid derived from the
disproportionation of magmatic SO2, as described by Reaction (24). At the high
temperature and low pH of this environment, isotopic equilibrium is expected among all
species and mineralogical sites (Rye et al. 1992). However, due to the elevated formational
temperature, the OH site is prone to retrograde isotopic exchange on cooling such that
intramineral SO4-OH fractionation and the hydrogen isotopic compositions may not be
representative of the conditions of formation for the alunite. Equilibrium fractionations
produce alunite sulfate with δ18O and δ34S that are dominantly a function of temperature
and the H2S/SO4 ratio of the hydrothermal fluid. Primary magmatic hydrothermal alunites
have SO4 δ18O values that are higher than those of supergene alunites and have OH δ18O
values that overlap those of the supergene alunite OH zone (Fig. 23). In reality, mixing
with local meteoric waters can produce SO4 values that overlap the supergene alunite SO4
field and OH zone, depending on the isotopic composition of meteoric water. Examples of
magmatic hydrothermal alunite occur in Peru, Colorado, Chile, Spain, Nevada, and
 Stable Isotope Systematics of Sulfate Minerals 581

elsewhere (Rye et al. 1992).

Magmatic steam alunites exhibit some similarities with both steam-heated and
magmatic hydrothermal alunites. However, magmatic steam alunite can be remarkably
coarse banded with zones of P2O5-enrichment variation and almost constant δ34S values
that reflect quantitative, disequilibrium oxidation of magmatic SO2 (and possibly H2S) by
entrainment of atmospheric oxygen or by loss of hydrogen from the hydrothermal fluid
(Cunningham et al. 1997). Their oxygen and hydrogen isotope characteristics are
dominated by equilibrium with magmatic water or derived steam; thus, the sulfate δ18O is
distinctly higher than that of the supergene alunite sulfate field (Fig. 23). The δ18O of the
OH is less distinctive because it lies within the supergene alunite OH zone, but the δD is
consistent with magmatic waters. Examples of magmatic steam alunite can be found in
Utah, California, and Colorado (Rye et al. 1992).
Stable isotope studies of simple sulfate minerals, such as anhydrite and barite, from
less acidic ancient porphyry and epithermal deposits are dominated by sulfur isotope
data, with fewer oxygen isotope data. Studies typically concentrate on the fractionation of
sulfur isotopes between coexisting sulfate and sulfide minerals, which can be useful for
constraining sources of sulfur, temperatures of mineralization, and equilibrium status of
mineralizing systems. Interpretations have utilized plots of the δ34S of coexisting sulfate
and sulfide minerals as a function of Δ sulfate−sulfide (Field and Gustafson 1976; cf. Shelton
and Rye 1982, Ohmoto 1986, Krouse et al. 1990). Theoretically, “δ-Δ” diagrams can
provide information about temperatures of mineralization and the SO4/H2S ratio of the
hydrothermal fluids provided that (1) the SO4/H2S ratio of the fluid remained constant,
(2) the bulk sulfur isotopic composition of the fluid (δ34SΣS) remained constant, and (3)
the only cause of isotopic variations in the initial δ34S was exchange between SO4 and
H2S in the fluid (Field and Gustafson 1976, Ohmoto 1986). Pairs of coexisting sulfate
and sulfide minerals plot along two linear arrays that converge to a point at Δ = 0 ‰,
which corresponds the δ34SΣS of the fluid. The angle between the linear arrays for the
sulfate and sulfide minerals can be related to the SO4/H2S ratio of the fluid (Field and
Gustafson 1976). However, because δ and Δ are not totally independent variables, linear
arrays can result from disequilibrium mineral pairs (Ohmoto 1986, Krouse et al. 1990). In
addition, natural settings rarely satisfy all of the conditions stated above. Interpretation of
natural data sets can be complicated by fluctuating fluid compositions, kinetic processes
related to isotopic exchange and precipitation, and mixing of multiple sulfur reservoirs,
among other processes (Shelton and Rye 1982, Ohmoto 1986, Krouse et al. 1990).
Similar δ-Δ diagrams have been interpreted in terms of equilibrium processes (Field
and Gustafson 1976) and in terms of disequilibrium processes (Shelton and Rye 1982).
Interpretations of equilibrium fractionation assume that the three conditions listed above
are satisfied. Interpretations of disequilibrium attribute the lack of equilibrium to non-
equilibrium oxidation-reduction in the fluid, mixing of fluids, dissolution and
reprecipitation of pre-existing sulfate and sulfide minerals, and (or) preservation of higher
temperature equilibria in the fluid (Shelton and Rye 1982). Shelton and Rye (1982)
proposed disequilibrium conditions at Mines Gaspé and El Salvador because anhydrite-
sulfide sulfur isotope temperatures were higher than temperatures derived from fluid
inclusion measurements. However, they used the sulfate-H2S fractionation factors from
Ohmoto and Rye (1979), which can yield temperatures that are in excess of 60°C greater
than those calculated using the sulfate-H2S fractionation of Ohmoto and Lasaga (1982;
Fig. 9). Thus, many of the anomalously high calculated temperatures may, in fact, be
within analytical error of the fluid inclusion temperatures if the Ohmoto and Lasaga
(1982) SO4H2S fractionation is used to recalculate temperature. Therefore, these systems
may be closer to equilibrium than previously thought.
582 Seal, Alpers & Rye

Figure 24. Plot of δ34S versus Δ SO 4 −Py for porphyry hydrothermal systems.
Values above the shaded fields are for sulfate minerals associated with the
sulfide minerals indicated in the legend; values below (and within) the shaded
field are for sulfide minerals. See text for sources of data. Shaded field shows
the permissive range for the bulk δ34S of the hydrothermal fluid for sulfide- and
sulfate- bearing porphyry systems. The solid lines indicate the hypothetical δ34S
values for sulfates (upper line) and sulfides (lower line) in equilibrium with a
hydrothermal fluid with equal proportions of SO4 and H2S. All isotopic values
in permil (VCDT).

The isotopic characteristics of sulfate and sulfide minerals from porphyry

environments suggest a general approach to equilibrium at elevated temperatures. Data
from El Salvador, Chile (Field and Gustafson 1976), Gaspé, Quebec (Shelton and Rye
1982), and Papua New Guinea (Eastoe 1983) are plotted in Figure 24. The data plot in
two linear trends, one for sulfate minerals, and one for sulfide minerals. The trends
converge with a y-intercept of ~ 3.5 ‰ and the trends are symmetrically distributed
above and below the intercept. The fractionations between SO4 and pyrite (Δ SO4 −Py ) range
from 6.6 to 18.2. Assuming equilibrium, these fractionations correspond to a temperature
range of 730 to 315°C, respectively. Regardless of equilibrium state, these data indicate
that the bulk δ34SΣS of the hydrothermal fluid lies between 0.5 and 6.5 ‰. Eastoe (1983)
questioned equating bulk fluid compositions with the composition of magmatic sulfur,
because of the complexities of the evolution of volatile phases from magmas. In high
temperature porphyry environments, Shelton and Rye (1982) suggested that the
discrepancy between fluid inclusion temperatures and SO4-sulfide sulfur isotope
temperatures resulted from the short time span between the disproportionation of SO2 to
SO42- and H2S, and the subsequent precipitation of sulfate as anhydrite.
Krouse et al. (1990) modeled the isotopic effects that would be expected from a
hydrothermal system undergoing active bacterial sulfate reduction. It was found that
converging trends for sulfates and sulfides, both with positive slopes, would be produced,
and it was suggested that this was a criterion indicating the possible involvement of
kinetic isotope effects. Krouse et al. (1990) also urged caution in the use δ vs. Δ
diagrams, and encouraged their interpretation in the context of other supporting studies
 Stable Isotope Systematics of Sulfate Minerals 583

such as petrographic observations of mineral assemblages, and independent estimates of

temperatures from fluid inclusions (cf. Shelton and Rye 1982, Eastoe 1983).
Sulfur isotope disequilibrium among aqueous sulfur species in ore-forming fluids
was recognized in large banded sulfide-barite veins in the Ag and base-metal ores of the
Creede mining district, Colorado (Rye and Ohmoto 1974). The barite shows a very large
range of both δ34S and δ18O values whereas the intergrown sulfides are characterized by a
narrow range of δ34S values (-2±2‰) typical of magmatic sulfur in a Tertiary volcanic
environment. The barite data fall in a three component mixing triangle (Fig. 25) with a
trend to higher δ34S and δ18O from north to south in the district (Rye et al. 1988, Barton
et al. 2000). The lower apex (δ18O = 2 ‰; δ34S = 1‰) in Figure 25 is interpreted as deep
sulfate from the hydrothermal system formed by the oxidation of H2S. The right apex
(δ18O = 20‰; δ34S = 35‰) requires that precursor aqueous sulfate went through a
bacteriogenic reduction cycle in a sedimentary environment, interpreted to have been the
sediments that formed the Creede Formation. The top apex δ18O = 8‰; δ34S = 48‰)
represents sulfate that evolved by bacteriogenic reduction but underwent further isotopic
evolution by oxygen isotope exchange with ore fluids and thermochemical reduction of
sulfate in the presence of organic matter. The isotope data on the barites at Creede are so
unusual for a hydrothermal deposit in a Tertiary volcanic environment that they indicated
some factor was present at Creede that does not normally exist in similar hydrothermal
systems. This unusual isotope data linked the origin of the ore fluids in the district to the
stratified saline lake (Rye et al. 2000a) that developed in the moat of the Creede caldera
and was a major factor leading to the choice of the Creede Moat by the Continental
Scientific Drilling Program (Bethke and Hay 2000).

Figure 25. Sulfur and oxygen isotope values

for hydrothermal barites from the Creede
mining district, Colorado (Rye et al. 1988,
Barton et al. 2000). All isotopic values in
permil (VSMOW and VCDT).

Mississippi Valley-type Pb-Zn-F deposits typically form in continental settings in low-

temperature (<200°C), near-neutral environments in which sulfur, and probably oxygen,
isotope disequilibrium would be expected to dominate (Ohmoto and Lasaga 1982, Chiba
and Sakai 1985). Thus, stable isotope data from sulfate minerals from Mississippi Valley-
type deposits should provide information about the source of sulfate and its geochemical
history. Stable isotope studies of Mississippi Valley-type deposits are dominated by sulfur
isotope data, with a limited amount of oxygen isotope data (Ault and Kulp 1960, Ohmoto
1986, Kaiser et al. 1987, Richardson et al. 1988, Kesler et al. 1994, Appold et al. 1995,
Kesler 1996, Misra et al. 1996, Jones et al. 1996). Most data are from barite, but some
sulfur isotope values have been reported for anhydrite, gypsum, and celestine (SrSO4).
Both sulfate sulfur and sulfide sulfur in Mississippi Valley-type environments can be
584 Seal, Alpers & Rye

derived from a variety of sources including organically bound sulfur, H2S reservoir gas,
evaporites, connate seawater, and diagenetic sulfides. In all cases, these sources represent
seawater sulfate that has followed various geochemical pathways which impart different
isotopic fractionations. The reduction of sulfate occurs either through bacterially mediated
processes or abiotic thermochemical processes. Bacterial sulfate reduction, as discussed
above, can produce sulfate-sulfide fractionations that typically range from 15 to 60 ‰
(Goldhaber and Kaplan 1975), whereas those associated with abiotic thermochemical
reactions with organic compounds range from nil to up 10 ‰ (Orr 1974, Kiyosu 1980).
Bacterial sulfate reduction has been documented at temperatures up to 110°C (Jørgensen et
al. 1992), but the optimum temperature range is between 30 and 40°C. Ohmoto and
Goldhaber (1997) argued that at the site of ore deposition, thermochemical reduction is not
effective at T <125°C because of slow reaction kinetics. For thermochemical reduction to
be an important process in Mississippi Valley-type environments, reduction must occur
away from the site of ore deposition, in the deeper, hotter parts of the basin. It should be
noted that although the kinetic fractionations associated with both reduction processes are
distinct, they can produce H2S of similar compositions if bacterial sulfate reduction occurs
quantitatively (or nearly so) in an environment with little or no Fe to sequester the sulfide.
Stable isotope and fluid-inclusion studies by Richardson et al. (1988) of samples from
the Deardorff mine from the Cave-in-Rock fluorspar district, Illinois, indicate
mineralization was dominated by two formation fluids recharged by meteoric waters, one
of which circulated into the basement rocks under low water/rock conditions. Liquid
hydrocarbons are present in fluid inclusion in most minerals. The low δ34S of sulfides (4.0
to 8.9 ‰ for sphalerite) indicates a significant contribution of H2S from petroleum sources.
These values are completely out of equilibrium with those for late stage barites which range
from about 57 to 103 ‰, with some crystals showing systematic increases from the core to
the edge of the crystals. Sulfate δ18O values however, range from only 19.6 to 20.8 ‰ and
were apparently in equilibrium with the hydrothermal fluids. These data suggest that the
aqueous sulfate was derived from a small fluid reservoir in which the residual seawater
sulfate underwent thermal chemical reduction with organic matter. Supporting evidence
includes the decrease in the δ13CCO2 of the fluids during carbonate deposition.
Sulfur isotope data from Mississippi Valley-type deposits suggest two major sulfide
reservoirs, one centered between -5 to 15 ‰ and one greater than 20 ‰ (Fig. 26). The
higher values of sulfides typically coincide with those of the composition of associated
sulfate minerals, and have been interpreted to reflect the minimal fractionation associated
with abiotic thermochemical reduction (Kesler 1996). However, similar compositions of
sulfide could be generated by closed-system, quantitative bacterial reduction of sulfate. A
carbonate aquifer is an ideal environment for such a geochemical process due to the lack of
reactive Fe to scavenge and fractionate sulfur. The lower values may reflect H2S derived
either directly or indirectly from open-system bacterial reduction of sulfate. Kesler et al.
(1994) proposed that low δ34S H2S was derived from oil in the deeper parts of the basin for
the Central Tennessee and Kentucky Mississippi Valley-type districts. This H2S ultimately
would have been derived from the bacterial reduction of sulfate. The H2S from both
bacterial and abiotic reduction is not in sulfur isotope equilibrium with associated sulfate
minerals (Fig. 26).
Oxygen isotope data for barites from the Southeast Missouri barite district mimics the
sulfur isotope data in defining hypogene (δ18O = 17.0 to 19.6 ‰, δ34S = 20.5 to 40.8 ‰)
and supergene fields (δ18O = 9.0 ‰, δ34S = 15.8 ‰; Kaiser et al. 1987). Because of the
extremely slow kinetics of oxygen and sulfur isotope exchange between SO4 and H2O and
H2S, respectively, at 100°C, the inferred temperature of mineralization, these compositions
undoubtedly reflect of the sources of the sulfate. The hypogene barite is probably
 Stable Isotope Systematics of Sulfate Minerals 585

Figure 26. Sulfur isotope histograms for Mississippi Yalley-type hydrothermal systems in
North America. Data from sphalerite are shown in black; data from sulfate minerals are
shown in white. All isotopic values in permil (VCDT).

dominated by sulfate ultimately derived from seawater, whereas the supergene barite
probably represents a mixture of dissolved hypogene barite, and sulfate generated by the
weathering of ore sulfide minerals.
Metamorphism of sulfate minerals
Metamorphism of assemblages containing a sulfate mineral and another sulfur-
bearing mineral, such as pyrite, may result in sulfur isotopic exchange. Isotopic re-
equilibration may involve both heterogeneous (mineral-mineral or mineral-fluid)
exchange reactions, and homogeneous (aqueous SO4 and H2S) exchange reactions. Key
aspects of the metamorphism of sulfate minerals are illustrated here with examples from
the Barite Hill gold deposit in the Carolina slate belt, South Carolina (Seal et al., in
press), the Åsen deposit, Skellefte district, Sweden (Rickard et al. 1979), and the Balmat-
Edwards Zn-Pb deposits, New York (Whelan et al. 1984).
At Barite Hill, barite and pyrite occur as both separate lenses and as intergrown
massive lenses hosted by metavolcanic and metasedimentary rocks and have been
interpreted to have formed by seafloor exhalative processes during Late Proterozoic time
(Seal et al., in press). Massive barite forms blades up to 3 cm in length and is associated
586 Seal, Alpers & Rye

with quartz. Grains of massive pyrite are generally less than 2 mm in diameter. In
intergrown assemblages, pyrite grains, less than 1 mm, are surrounded by barite, quartz,
and silicate gangue. The sulfur isotope compositions of the barite and pyrite vary
systematically with sulfur-bearing mineral assemblages. The δ34S values of massive
barite cluster tightly between 25.0 and 28.0 ‰, and are almost identical to Late
Proterozoic seawater values; the δ34S of massive sulfide ranges from 1.0 to 5.3 ‰, and is
consistent with seafloor hydrothermal systems. In contrast, the δ34S values of finer-grain,
intergrown pyrite (5.1 to 6.8 ‰) and barite (21.0 to 23.9 ‰) are higher and lower than
their massive counterparts, respectively (Fig. 27). The area of sulfur isotope exchange
appears to be on the scale of several millimeters, because massive barite within 1 or 2 cm
of barite intergrown with pyrite has δ34S values that are 3 to 5 ‰ higher than the barite
intergrown with pyrite. Intergrown pairs of barite and pyrite yield temperatures from 332
to 355°C, which is consistent with the Ordovician greenschist facies regional
metamorphism that affected the deposit (Fig. 28).

Figure 27. Sulfur isotope histogram for barite and pyrite from the Barite Hill
deposit, South Carolina. Modified from Seal et al. (in press). Black symbols are
for sulfide minerals; open symbols are for barite. Shaded field represents typical
seafloor hydrothermal sulfide compositions; diagonally dashed fields represents
Late Proterozoic seawater composition. All isotopic values in permil (VCDT).

At Balmat-Edwards, anhydrite is associated with sphalerite, pyrite, and galena in

massive lenses hosted by marble (Whelan et al. 1984). The entire package was
metamorphosed to amphibolite facies conditions (~600°C and ~6.5 kb) at approximately
1 Ga (Bohlen et al. 1985). The δ34S of ore sulfide minerals ranges from 11.5 to 17.5 ‰,
whereas that of associated anhydrite ranges from 25.6 to 27.0 ‰. The δ34S of evaporitic
anhydrite and associated pyrite stratigraphically above the orebodies shows a wider range
from 4.2 to 30.2 ‰, and from -31.6 to 20.1 ‰, respectively (Whelan et al. 1984, 1990).
Using the fractionation factors in Table 1, anhydrite-pyrite pairs from both the ores and
the evaporites yield sulfur isotope temperatures that range from 504 to 595°C (excluding
one temperature of 411°C), which are consistent with peak metamorphic conditions,
possibly modified by minor amounts of retrograde re-equilibration (Fig. 28). In contrast,
temperatures based on sulfur isotopic fractionations for sulfide-mineral pairs (pyrite-
sphalerite, sphalerite-galena, pyrite-galena) are much lower, ranging from 251 to 597°C,
and have been interpreted to reflect either retrograde re-equilibration to lower
temperatures, or a second, lower grade metamorphic event that was not sufficiently
intense to reset anhydrite-pyrite pairs (Whelan et al. 1984).
The study of Rickard et al. (1979) of the Åsen deposit in the Skellefte district of
Sweden illustrates the potential pitfalls associated with the use of inaccurate fractionation
factors. The Åsen deposits consist of massive stratiform and stratabound massive pyrite +
 Stable Isotope Systematics of Sulfate Minerals 587

Figure 28. Plot of δ34S for sulfate

minerals versus δ34S for pyrite from
metamorphosed mineral deposits.
Peak metamorphic conditions for
Balmat are shown in shaded field.
Greenschist conditions are shown as
stippled field. The change in δ34S for
barite at Barite Hill is shown by the
arrows going from massive barite (the
presumed starting composition for the
intergrown barite) to the intergrown
barite. All isotopic values in permil
(VCDT).

barite lenses, hosted by 1.8 Ga submarine volcanic rocks. Rickard et al. (1979) noted a
correlation between δ34S values for barite and pyrite and found that the barite-pyrite sulfur
isotopic fractionation (Δ barite−pyrite ) ranged from 11.7 to 17.1 ‰ with a mean of 13.8 ‰. On
the basis of the older sulfate-pyrite fractionation factors (Ohmoto and Rye 1979), a mean
apparent temperature of470°C was calculated. This was concluded to be too high for both
the greenschist-facies metamorphic and hydrothermal histories of the deposits; a
disequilibrium model was therefore formulated to explain the observed isotopic
compositions. The model included the oxidation of volcanic sulfur and bacterial sulfate
reduction. However, on the basis of more recent SO4-H2S fractionation factors (Ohmoto
and Lasaga 1982), the sulfur isotope data of Rickard et al. (1979) yield temperatures
between 332 and 465°C (mean = 404°C), which are consistent with greenschist-facies
metamorphism (Fig. 28). Nevertheless, many aspects of the Rickard et al. (1979)
disequilibrium model for Åsen are necessary to explain the sulfur isotopic variations.
However, the correlation between barite and pyrite δ34S values probably reflects a
metamorphic overprint.
Isotopic re-equilibration of sulfate minerals during metamorphism seems to be
dependent on a variety of kinetic factors, which probably includes grain size, presence or
absence of a fluid phase, presence or absence of recrystallization during metamorphism,
and retrograde vs. prograde solubility. Given the appropriate circumstances, isotopic re-
equilibration of sulfate minerals seems to occur at metamorphic grades at least as low as
greenschist facies.
Surficial environments
Continental evaporites, saline lakes, and ground waters. Stable isotope studies of
saline lakes and non-marine evaporites include investigations of Late Cretaceous to Early
Tertiary evaporites in Bolivia (Rouchy et al. 1993), Tertiary evaporites in Spain
(Birnbaum and Coleman 1979, Utrilla et al. 1992), recent evaporitic gypsum deposits in
Namibia (Eckardt and Spiro 1999), modern playa lakes in Australia (Chivas et al. 1991,
Alpers et al. 1992), and Searles Lake in California (Holser and Kaplan 1966). The data
are dominantly from gypsum, and anhydrite, with some data coming from alunite,
jarosite, epsomite MgSO4·7H2O, celestine SrSO4, blödite Na2Mg(SO4)2·4H2O, mirabilite
Na2SO4·10H2O, thenardite Na2SO4, hanksite KNa22(SO4)(CO3)2Cl, sulphohalite
Na6(SO4)2FCl, aphthitalite (K,Na)3Na(SO4)2, and burkeite Na6(CO3)(SO4)2.
588 Seal, Alpers & Rye

Non-marine evaporites typically form in continental interiors in rain shadows of major

mountain ranges, such as the Great Basin of Nevada, California, Utah, Idaho, Colorado,
and northern Arizona in the rain shadow of the Cordillera. The stable isotope characteristics
of sulfate from saline lakes and continental (non-marine) evaporites reflect sources and
processes within the local catchments in addition to regional, climatic influences. Sulfate in
these environments typically is derived from a variety of sources, which may include: (1)
atmospheric deposition of marine aerosols; (2) dissolution of pre-existing evaporites in the
basin; (3) marine salts from marine sedimentary rocks (connate salts); (4) “relict” sea salts
left by oceanic regression; (5) evaporated river water; and (6) weathering of sulfide
minerals in the basement rocks of the basin (Chivas et al. 1991). Marine aerosols consist of
both sea-spray sulfate, with δ34S= 21.0 ‰ for modern spray (the composition of seawater),
and “non-sea-salt” sulfate (NSS). NSS sulfate is dominantly derived from the atmospheric
oxidation of organic sulfur compounds such as dimethylsulfide (CH3SCH3; DMS), that is
produced by marine organisms. Additional non-sea-salt sulfate can be derived from the
oxidation of H2S produced by bacterial sulfate reduction in marine sediments.
Dimethylsulfide from the central Pacific has δ34S = l5.6±3.l ‰ (Calhoun et al. 1991). Bulk
stratospheric sulfur, which includes oxidized DMS and H2S, among other compounds, has
δ34S = 2.6±0.3 ‰ when not perturbed by volcanic inputs (Castleman et al. 1974). As
discussed above, H2S produced from bacterial reduction of modern seawater sulfate should
have δ34S = -20±20 ‰. Sea-spray droplets are expected to be deposited from rainfall in
coastal regions and from low altitude air masses, whereas oxidized biogenic sulfur
compounds (i.e. DMS) reach higher altitudes, and consequently are expected to precipitate
farther inland (Wakshal and Nielsen 1982, Chivas et al. 1991).
The relative importance of the different sulfate sources varies with study area (Fig.
29). In Australia, the sulfate budget of the non-marine playas appears to be dominated by
marine aerosols. Near the coast, the δ34S is identical to modern seawater values (δ34S ≈ 21
‰), reflecting the sea-salt-spray component, whereas inland, values systematically range
down to approximately 15 ‰, which has been interpreted to reflect greater inputs from
higher altitude, oxidized DMS (Chivas et al. 1991). More negative, sedimentary basement
sources are only locally important. In Namibia, sulfate in gypsum is dominantly derived
from oxidized atmospheric DMS, primarily produced offshore in the Benguela Current;
oxidation of sulfide-rich ore bodies is only locally important to the sulfate budget, resulting
in gypsum δ34S values of approximately 3 ‰ (Fig. 29; Eckardt and Spiro 1999). In Spain,
sulfate in non-marine evaporites was recycled from Triassic and, to a lesser extent,
Cretaceous marine evaporites exposed in the basins. Minor modification of isotopic
signatures has been attributed to redox cycling associated with bacterial sulfate reduction
(Birnbaum and Coleman 1979, Utrilla et al. 1992). Sulfate in Bolivian evaporites was
derived from a combination of recycled sulfate in Permo-Triassic marine evaporites and
weathering of igneous sulfide minerals in the basins, and may have locally been modified
by bacterial sulfate reduction (Rouchy et al. 1993). Sulfate in Searles Lake, California, has
been interpreted to be derived from a combination of igneous and atmospheric sources,
with subsequent modification by bacterial sulfate reduction (Holser and Kaplan 1966).
The δ18O of sulfate from these systems in Spain (Utrilla et al. 1992), and Bolivia
(Rouchy et al. 1993) ranges from 11.3 to 19.7 ‰, with one value of-0.5 ‰. These values
generally reflect the same marine evaporite sources and bacterial sulfate-reduction processes
as do the δ34S data. The δ18O of alunite and jarosite from acidic, hypersaline Lake Tyrrell in
Victoria, Australia, ranges from 22.5 to 24.9 ‰, and represents some of the highest values
reported for natural sulfates (Alpers et al. 1992). Alpers et al. (1992) suggested that the
anomalously high δ18O was the result of oxygen isotopic exchange between dissolved SO4
and H2O in the ground-water aquifers that feed Lake Tyrrell; they proposed that the δ18O
values of the SO4 and H2O reflect between 63 and 89 % equilibration. The exchange was
 Stable Isotope Systematics of Sulfate Minerals 589

Figure 29. Histogram of δ34S values for sulfate minerals from non-marine
evaporites, and saline lakes. Marine DMS field is from Calhoun et al.
(1991). Biogenic and sedimentary sulfide at Searles Lake has an inferred
δ34S below -20 ‰. The ranges of δ34S for Triassic and Tertiary marine
sulfate sources in the Erbo basin are shown by the arrows. All isotopic
values in permil (VCDT).

facilitated by long residence times of water, and the enhanced kinetics at the acidic pH (2.9
to 4.0) of the ground waters. Using kinetic data from Lloyd (1968), Alpers et al. (1992)
suggested that the equilibration state of the SO4-H2O oxygen isotopes indicated minimum
residence times of 110 to 670 yr at 25°C for sulfate in the regional ground waters.
However, extrapolation of the kinetic data of Hoering and Kennedy (1957) and Chiba and
Sakai (1985) to pH = 3.0 to 4.0 at 25°C indicates significantly longer (and probably
unrealistic) residence times of 17,000 to 400,000 yr to achieve 63 to 89 % equilibrium.
Acid mine drainage. The stable isotope characteristics of acid mine drainage have
been reviewed by van Everdingen and Krouse (1988) and Taylor and Wheeler (1994).
Dissolved sulfate concentrations can typically range up to 5 g/L, but extreme
concentrations up to 760 g/L have been reported where evaporation is important
(Nordstrom et al. 2000). In evaporative settings, mine drainage can precipitate a variety of
secondary, efflorescent sulfate salts, such as gypsum, chalcanthite CuSO4·5H2O,
590 Seal, Alpers & Rye

melanterite FeSO4·7H2O, rozenite FeSO4·4H2O, szomolnokite FeSO4·H2O, copiapite

FeIIFeIII4(SO4)6(OH)2·20H2O, and halotrichite FeIIAl2(SO4)4·22H2O, to name a few (Alpers
et al. 1994, Nordstrom and Alpers 1999b).
Sulfate in these systems is dominantly derived from the oxidation of pyrite and other
sulfide minerals. The oxidation of pyrite is described by Reactions (20) and (21), which
differ in terms of oxidizing agents and sources of oxygen for the resulting sulfate, and
impart distinctly different oxygen isotopic signatures, as discussed above. The δ34S of
dissolved SO4 is typically identical to that of the parent sulfide mineral, because of
quantitative, disequilibrium oxidation (Field 1966, Taylor et al. 1984, Taylor and
Wheeler 1994). The δ18O of dissolved sulfate depends on the composition of local water
and the mechanism and environment of pyrite oxidation (Toran and Harris 1989, Taylor
and Wheeler 1994).
Taylor and Wheeler (1994) discussed the application of plots δ18OSO4 versus δ18O H2O
to the interpretation of the stable isotope characteristics of mine drainage. Among the
several caveats and qualifiers that bear on the applicability and limitations of this approach,
are the assumption that the isotopic fractionations are constant, that isotopic re-
equilibration subsequent to the initial formation of sulfate has not occurred, and that
oxygen sources participate in a fully molecular pathway in which atoms of oxygen are
transferred. However, oxygen may be involved in electron transfer in a strictly
electrochemical pathway. In the use of δ18OSO4 versus δ18O H2O plots, the oxidation of pyrite
through Reactions (20) and (21) is represented by graphical depictions of Equation (23),
where the percentages of sulfate contribution from Reaction (20) and Reaction (21) are
shown (Fig. 14). In Reaction (20), pyrite is oxidized by dissolved oxygen, and 87.5 % of
the oxygen is contributed to the resulting sulfate from dissolved oxygen and 12.5 % is
derived from water. In Reaction (21), dissolved Fe3+ oxidizes pyrite and all of the oxygen
in the resulting sulfate is derived from water. Samples that plot near Reaction (20) (0 %
isopleth), and consequently have higher Δ SO4 −H2O values, are interpreted to represent
unsaturated (or wet/dry), oxygenated environments. Samples that plot near Reaction (21)
(100 % isopleth), with lower Δ SO4 −H2O values, are interpreted to represent submersed, less
oxic environments (Fig. 14). The 2.7 ‰ difference in fractionation between dissolved
oxygen and sulfate between acidic ( ε O2 = -11.4 ‰) and neutral to alkaline ( ε O2 = -8.7 ‰)
conditions causes minor changes to the locations of isopleths for Reactions (20) and (21) as
can be seen from Equation (23).
Stable isotope data are available for the West Shasta district, the Leviathan mine and the
Penn mine, California, the Argo Tunnel, Colorado, the Fontana and Hazel Creek mines,
North Carolina, and the Sulfur and Cabin Branch mines, Virginia (Fig. 14; Taylor and
Wheeler 1994, Hamlin and Alpers 1996, Seal and Wandless 1997). All of these sites are
base-metal sulfide deposits, where the primary sulfide minerals are pyrite, or pyrrhotite, with
lesser chalcopyrite, sphalerite, galena, and native sulfur. Collectively, the data from these
studies illustrate several important points regarding the interpretation and applicability of
δ18OSO4 versus δ18O H2O diagrams. The data from these sites spans a range of δ18OSO4 and
δ18O H2O . The range of δ18O H2O values reflects the temperature dependence of δ18O the of
meteoric waters, which can be related to variations in altitude and latitude, and the effect of
evaporation, which raises the δ18O of residual water (Fig. 1). The range of δ18OSO4 values
dominantly reflects the environment of sulfide oxidation. The data for the Leviathan and
Sulfur mines, and the Argo Tunnel suggest that Fe3+ (Reaction 21) is responsible for greater
than 50 % of the sulfide oxidation from these sites, which is consistent with the water
saturated nature of much of the mine workings at these sites. In the case of the Fontana and
Hazel Creek mines, and parts of the West Shasta district, the data indicate that dissolved
oxygen is responsible for between 50 and 75 % of the sulfide oxidation, which is consistent
 Stable Isotope Systematics of Sulfate Minerals 591

with the location of some parts of these deposits in the unsaturated zone.
Several processes, such as evaporation and dissolution of efflorescent salts, complicate
the interpretation of data using these diagrams. Evaporation will shift the δ18O of waters to
more positive values without affecting the δ18O of the sulfate. The importance of
evaporation at a site can be assessed from the δD and δ18O signatures of the waters;
evaporation produces water that has higher δD and δ18O values than local meteoric water.
Evaporation is known to be an important process at the Penn, Sulfur, and Cabin Branch
mines (Hamlin and Alpers 1996, Seal and Wandless 1997). The isotopic composition of
efflorescent salts should be representative of the environment of initial sulfide oxidation.
However, when the salts redissolve, the δ18O of the waters may or may not be
representative of the conditions of initial sulfide oxidation, because of seasonal variations
in the δ18O of meteoric waters due to differences in the temperature of precipitation. Thus,
the δ18O of water associated with dissolved efflorescent salts may be higher or lower than
the δ18O of water that originally carried the sulfate, potentially leading to erroneous
interpretations. In the case of the West Shasta district, the concentrations of some of the
waters from Iron Mountain are the most extreme ever recorded, with pH values as low as -
3.6, SO4 concentrations up to 760 g/L, dissolved metal concentrations in excess of 160 g/L,
and temperatures up to 47°C (Nordstrom et al. 2000). These conditions enhance the
kinetics of oxygen isotope exchange between SO4 and H2O. At 50°C, SO4 in equilibrium
with a water with δ18O = -10 ‰ should have δ18O = 15.4 ‰. Alpers et al. (1996) reported
δ18O values in aqueous sulfate and efflorescent salts from Iron Mountain ranging from 10
to 15 ‰, indicating partial to total isotopic equilibration with the highly acidic mine waters.
Extrapolation of the data from Hoering and Kennedy (1957) and Chiba and Sakai (1985)
indicates that an approach to 60 % ofisotopic equilibrium can be achieved in only 15 hours
at pH = -4 and 1.5 years at pH = 0 (Fig. 12).
Paleoclimate studies. Stable isotope studies of sulfate minerals have received limited
application to paleoclimate studies. Arehart and O’Neil (1993) investigated the
continental paleoclimate record of supergene alunites from Nevada, and Khademi et al.
(1997) investigated the paleoclimate associated Iranian gypsums. Alunite and jarosite are
ideally suited for paleoclimate studies; K within the structure is amenable to K/Ar and
40
Ar/39Ar dating and clear stable isotope criteria exist for distinguishing supergene origins
alunite and jarosite from other origins (Rye et al. 1992, Arehart et al. 1992, Arehart and,
O’Neil 1993, Rye and Alpers 1997). Arehart and O’Neil (1993) used the δD of supergene
alunites as an indicator of ancient meteoric waters over the past 30 Ma in Nevada; the
results were then related to temperature variations due to the changing temperatures of
precipitation. There appears to be little or no hydrogen isotopic fractionation between
alunite and water at ambient conditions (Bird et al. 1989). Arehart and O’Neil (1993)
were able to document changes in the δD compositions of ancient meteoric waters, and
by inference, mean annual temperatures of precipitation, that ranged to higher and lower
values than present-day conditions (Fig. 30).
Rye et al. (2000b) used a combination of isotopic ages and stable isotope data on
supergene alunite and jarosite formed by the weathering of the Creede deposits to trace the
evolution of the chemical and hydrologic processes that affected ancient oxidized acid
ground waters, as well as the details of climate history and geomorphic evolution in the San
Juan Mountains. Fine grained (>1μm to <10 μm) supergene alunite and jarosite occur in
minor fractures in the upper, oxidized parts of the 25 Ma sulfide-bearing veins of the
Creede mining district, and jarosite also occurs in adjacent oxidized Ag-bearing clastic
sediments. K/Ar ages for alunite range from 4.8 to 3.1 Ma, and for jarosite from 2.6 to 0.9
Ma. The δD values for alunite and jarosite shows opposite correlations with elevation, and
values for jarosite correlate with age. Calculated δD H 2O values of alunite fluids approach,
592 Seal, Alpers & Rye

Figure 30. Secular variations in the δD of supergene alunites from Nevada.

Modified from Arehart and O’Neil (1993). White values with arrows are
corrected for latitude (Arehart and 0’Neil 1993). PDMW indicates present-
day meteoric water compositions. More negative values are inferred to
represent cooler climates, and less negative values are inferred to represent
warmer climates. All isotopic values in permil (VSMOW).

but are higher than, those of present-day meteoric water. Calculated δD H 2O values for
jarosite fluids are more variable; the values of the youngest jarosite are lowest and are
similar to those of present day meteoric water in the district. The narrow range for δD-
δ18OSO4 values of alunite reflects oxidation of sulfide below the water table. The greater
range in these values for jarosite reflects oxidation of sulfide under vadose conditions. Ages
of alunite can be used to mark the position of the paleo-water table at the end of a period of
moderate erosion from 25 to 5 Ma that exposed the tops of the ore-bodies to oxidation. The
younger jarosite formed in the vadose zone during or following canyon cutting related to
regional uplift of the southern Rocky Mountains. The δD values suggest that climate in the
area was similar to that of the present day prior to regional uplift, but went through a warm
period before returning to present conditions during or after regional uplift.
Khademi et al. (1997) measured the δD and δ18O of hydration water in gypsum to
gain insights into the waters associated with gypsum formation. Their data demonstrated
that evaporation of brines was the most important process resulting in gypsum
precipitation. It was also found that an evaporation trend for the inferred “mother” waters
intersected the meteoric water line at compositions lower than modern meteoric water
compositions. This fact was used to suggest that climates in the Quaternary were more
moist and (or) cooler than modern climates, or that Quaternary climate patterns were
dramatically different from modern patterns.

SUMMARY
The stable isotope characteristics of dissolved sulfate and sulfate minerals are typically
dominated by the kinetic rates of exchange reactions among sulfate, H2O, and H2S,
although other less abundant species may also be important. The oxidation state of sulfur
species plays a dominating role in causing large sulfur isotope fractionations, with oxidized
species generally having higher δ34S values than reduced species. The kinetic rates of the
exchange reactions are strongly dependent on temperature and pH. At high temperature and
 Stable Isotope Systematics of Sulfate Minerals 593

low pH, equilibrium dominates and results from abiotic processes. Due to slower kinetics,
high-temperature isotopic signatures are usually preserved during cooling. Thus, stable
isotope data commonly provide a record of ambient conditions, such as temperature, and
fluid compositions.
In contrast, at lower temperature and less acidic pH, disequilibrium, mediated by
microbial processes, prevails. Under these conditions, stable isotope data tend to record
information about the source of components and the nature of (bio)geochemical
processes acting on the system. Isotopic differences among the various geologically
important, terrestrial sulfur reservoirs arose as a result of redox-dependent fractionation
of the bulk Earth sulfur early in geological time. Under appropriate redox conditions,
sulfate-reducing bacteria and iron- and sulfur-oxidizing bacteria played important roles,
which are inadequately understood at present. Disequilibrium processes are responsible
for the secular variations observed in the marine record. Reduced sulfur from the bulk
Earth is the major reservoir, followed by seawater sulfate.
Key elements of our knowledge that appear to have the most significant impact on
the interpretation of the stable isotope geochemistry of sulfate include revisions to our
understanding, of SO4-H2S sulfur isotope fractionations (Ohmoto and Lasaga 1982), and
persisting uncertainties in the rates of oxygen isotope exchange between dissolved SO4
and H2O. Discrepancies between temperatures calculated from the δ34S of sulfate-sulfide
pairs using previous-fractionation factors versus those of Ohmoto and Lasaga (1982)
range from nil at 200°C to more than 50°C at temperatures above 300°C (Fig. 9). This
discrepancy has led several researchers to formulate elaborate geochemical
interpretations because calculated sulfur-isotope temperatures could not be reconciled
with other geological and geochemical constraints.
The temperature and pH dependence of oxygen isotope exchange rates between
dissolved sulfate and water from the experimental studies of Lloyd (1968) and Chiba and
Sakai (1985) are in stark contrast to one another; the rates of Lloyd (1968) are several
orders of magnitude faster than those of Chiba and Sakai (1985). Probably because of
their earlier publication date, the data from Lloyd (1968) have served as a foundation for
much of the interpretation of oxygen isotope data from sulfates. However, recalculation
of the experimental fluid compositions of Hoering and Kennedy (1957) in terms of pH by
the present study permits comparisons with the data of Lloyd (1968) and Chiba and Sakai
(1985). Agreement found between the studies of Hoering and Kennedy (1957) and Chiba
and Sakai (1985) suggests, although not conclusively, that the exchanges rates from these
studies are more accurate than those of Lloyd (1968). Accordingly, many of the
fundamental aspects of the interpretation of the oxygen isotope geochemistry of sulfates
may need to be re-evaluated.
ACKNOWLEDGMENTS
The preparation of this manuscript benefited from discussions with JK Böhlke, Tyler
Coplen, Wayne Goodfellow, Carol Ptacek, Pat Shanks, and Rob Zierenberg. The
manuscript also benefited from reviews by Phil Bethke and Jane Hammarstrom. Adam
Johnson and Greg Wandless assisted in literature searches.
REFERENCES
Alpers CN, Nordstrom DK, White LD (1988) Solid solution properties and deuterium fractionation factors for
hydronium-bearing jarosites from acid mine waters. EOS, Trans Am Geophys Union 69:1480-1481
Alpers CN, Rye RO, Nordstrom DK (1996) Stable isotope systematics of S and O in aqueous and mineral
sulfates from hyper-acid environments [abstr]. Chapman Conference on crater lakes, terrestrial degas-
sing, and hyper-acid fluids in the environment. Crater Lake, Oregon, Sept. 4-9, 1996, Am Geophys
Union, Washington, DC
594 Seal, Alpers & Rye
Alpers CN, Rye RO, Nordstrom DK, White LD, King B-S (1992) Chemical, crystallographic and stable
isotopic properties of alunite and jarosite from acid-hypersaline Australian lakes. Chem Geol 96:203-
226
Alpers CN, Blowes DW, Nordstrom DK, Jambor JL (1994) Secondary minerals and acid mine drainage. In
Environmental Geochemistry of Sulfide Mine-Wastes. Jambor JL, Blowes DW (Eds) Mineral Assoc
Canada Short Course Handbook 22:247-270
Alt JC, Anderson TF, Bonnell L (1989) The geochemistry of sulfur in a 1.3-km section of hydrothermally
altered oceanic crust, DSDP 504B. Geochim Cosmochim Acta 53:1011-1023
Appold MS, Kesler SE, Alt JC (1995) Sulfur isotope and fluid inclusion constraints on the genesis of
Mississippi Valley-type mineralization in the Central Appalachians. Econ Geol 90:902-919
Arehart GB, O’Neil JR (1993) D/H ratios of supergene alunite as an indicator of paleoclimate in
continental settings. In Climate change in continental isotopic records. Swart PK, Lohmann KC,
McKenzie JA, Savin S (Eds) Geophys Monogr 78:277-284
Arehart GB, Kesler SE, O’Neil JR, Foland KA (1992) Evidence for the supergene origin of alunite in
sediment-hosted micron gold deposits, Nevada. Econ Geol 87:263-270
Ault WU, Kulp JL (1960) Sulfur isotopes and ore deposits. Econ Geol 55:73-100
Ault WU, Jensen ML (1963) Summary of sulfur isotope standards. In Biogeochemistry of Sulfur Isotopes.
Jensen ML (Ed) Nat Sci Foundation, Symp Proc, Yale University, New Haven, CT
Bachinski DJ (1969) Bond strength and sulfur isotopic fractionation in coexisting sulfides. Econ Geol
64:56-65
Baertschi P (1976) Absolute 18O content of Standard Mean Ocean Water. Earth Planet Sci Lett 31:341-344
Banks NG (1982) Sulfur and copper in magma and rocks. In Advances in Geology of the Porphyry Copper
Deposits in Southwestern North America. Titley SR (Ed) University of Arizona Press, Tucson,
Arizona, p 227-258
Barton PB, Rye RO, Bethke PM (2000) Evolution of the Creede caldera and its relation to mineralization in
the Creede mining district, Colorado. In Ancient Lake Creede, its volcano-tectonic setting,
sedimentation history, and relation to mineralization in the Creede mining district. Bethke PM, Hay
RL (Eds) Geol Soc Am Spec Paper 346:301-326
Bahr JR (1976) Sulfur isotopic fractionation between H2S, SO42- and in aqueous solutions and possible
mechanisms controlling isotopic equilibrium in natural systems. MSc thesis, Pennsylvania State
University
Bernard A, Demaiffe D, Mattielli N, Punongbayan RS (1991) Anhydrite-bearing pumices from Mount
Pinatubo: Further evidence for the existence of sulphur-rich silicic magmas. Nature 354:139-140
Berner RA (1985) Sulphate reduction, organic matter decomposition and pyrite formation. Phil Trans R
Soc London A 315:25-38
Bethke PM, Hay RL (2000) Overview: Ancient Lake Creede, its volcano-tectonic setting, sedimentation
history, and relation to mineralization in the Creede mining district. In Ancient Lake Creede, its
volcano-tectonic setting, sedimentation history, and relation to mineralization in the Creede mining
district. Bethke PM, Hay RL (Eds) Geol Soc Am Spec Paper 346:1-8
Betts RH, Voss RH (1970) The kinetics of oxygen exchange between sulfite ion and water. Canad J Chem
48:2035-2041
Bigeleisen J (1952) The effects of isotopic substitution on the rates of chemical reactions. J Phys Chem
56:823-828
Bigeleisen J, Mayer MG (1947) Calculation of equilibrium constants for isotopic exchange reactions. J
Chem Phys 15:261-267
Bigeleisen J, Perlman ML, Prosser HC (1952) Conversion of hydrogenic materials for isotopic analysis.
Anal Chem 24: 1356
Bird MI, Andrew AS, Chivas AR, Lock D (1989) An isotopic study of surficial alunite in Australia: 1.
Hydrogen and sulfur isotopes. Geochim Cosmochim Acta 53:3223-3238
Birnbaum SJ, Coleman M (1979) Source of sulphur in the Ebro basin (northern Spain) Tertiary nonmarine
evaporite deposits as evidenced by sulphur isotopes. Chem Geol 25:163-168
Bischoff JL, Seyfried WE (1978) Hydrothermal chemistry of seawater from 25° to 350°C. Am J Sci
278:838-860
Bluth GJS, Rose WI, Sprod IE, Krueger AJ (1997) Stratospheric loading of sulfur from explosive volcanic
eruptions. J Geol 105:671-683
Bohlen SR, Valley JW, Essene EJ (1985) Metamorphism in the Adirondacks. 1. petrology, pressure, and
temperature. J Petrol 26:971-992
Borthwick J, Harmon RS (1982) A note regarding CIF3 as an alternative to BrF5 for oxygen isotope
analysis. Geochim Cosmochim Acta 46:1665-1668
Brenninkmeijer CAM, Kraft P, Mook WG (1983) Oxygen isotope fractionation between CO2 and H2O.
Isotope Geosci 1:181-190
 Stable Isotope Systematics of Sulfate Minerals 595
Burnham CW, Ohmoto H (1980) Late-stage process of felsic magmatism. Soc Mining Geol Japan Spec
Issue 8:1-11
Ҫagatay MN, Eastoe CJ (1995) A sulfur isotope study of volcanogenic massive sulfide deposits of the
Eastern Black Sea province, Turkey. Mineral Deposita 30:55-66
Calhoun JA, Bates TS, Charlson RJ (1991) Sulfur isotope measurements of submicrometer sulfate aerosol
particles over the Pacific Ocean. Geophys Res Lett 18:1877-1880
Carmody RW, Plummer LN, Busenberg E, Coplen TB (1998) Methods for collection of dissolved sulfate
and sulfide and analysis of their sulfur isotopic composition. US Geol Survey Open-File Report 97-
234
Carroll MR, Rutherford MJ (1987) The stability of igneous anhydrite: Experimental results and
implications for sulfur behavior in the 1982 El Chichón trachyandesite and other evolved magmas. J
Petrol 28:781-801
Carroll MR, Rutherford MJ (1988) Sulfur speciation in hydrous experimental glasses of varying oxidation
state: results from measured wavelength shifts of sulfur X-rays. Am Mineral 73:845-849
Castleman AW Jr, Munkelwitz HR, Manowitz B (1974) Isotopic studies of the sulfur component of the
stratospheric aerosol layer. Tellus 26:222-234
Chambers LA, Trudinger PA (1979) Microbiological fractionation of stable sulfur isotopes: A review and
critique. Geomicrobiol J 1:249-293
Chiba H, Sakai H (1985) Oxygen isotope exchange rate between dissolved sulfate and water at
hydrothermal temperatures. Geochim Cosmochim Acta 49:993-1000
Chiba H, Kusakabe M, Hirano S-I, Matsuo S, Somiya S (1981) Oxygen isotope fractionation factors
between anhydrite and water from 100 to 550°C. Earth Planet Sci Lett 53:55-62
Chivas AR, Andrew AS, Lyons WB, Bird MI, Donnelly TH (1991) Isotopic constraints on the origin of
salts in Australian playas: 1. Sulphur. Palaeogeog Palaeoclim Palaeoecol 84:309-332
Claypool GE, Holser WT, Kaplan IR, Sakai H, Zak I (1980) The age curves of sulfur and oxygen isotopes
in marine sulfate and their mutual interpretations. Chem Geol 28:199-260
Clayton RN (1986) High temperature effects in the early solar system. Rev Mineral 16:129-140
Clayton RN, Mayeda TK (1963) The use of bromine pentafluoride in the extraction of oxygen from oxides
and silicates for isotopic analysis. Geochim Cosmochim Acta 27:43-52
Cole DR, Ohmoto H (1986) Kinetics of isotopic exchange at elevated temperatures and pressures. Rev
Mineral 16:41-90
Coleman ML, Moore MP (1978) Direct reduction of sulfates to sulfur dioxide for isotopic analysis. Anal
Chem 50:1594-1595
Coleman ML, Sheppard TJ, Durham JJ, Rouse JE, Moore GR (1982) Reduction of water with zinc for
hydrogen isotope analysis. Anal Chem 54:993-995
Cortecci G (1974) Oxygen isotopic ratios of sulfate ions-water pairs as a possible geothermometer.
Geothermics 3:60-64.
Craig H (1961) Isotopic variations in meteoric waters. Science 133:1702-1703.
Craig H, Gordon LI (1965) Deuterium and oxygen-18 variations in the ocean and marine atmosphere.
Symposium on marine geochemistry, Narragansett Marine Laboratory, Univ Rhode Island Publ
3:277-374
Criss RE, Taylor HP Jr. (1986) Meteoric-hydrothermal systems. Rev Mineral 16:373-424
Cunningham CG, Rye RO, Bethke PM, Logan MAV (1997) Vein alunite formed by degassing of an
epizonal stock. Eos, Trans 78:F746
Dansgaard W (1964) Stable isotopes in precipitation. Tellus 16:436-468
Dansgaard W, Johnsen SJ, Møller J, Langway CC Jr. (1969) One thousand centuries of climatic record
from Camp Century on the Greenland ice sheet. Science 166:377-381
Deen, JA (1990) Hydrothermal ore deposition related to high-level igneous activity: a stable isotope study
of the Julcani mining district, Peru. Unpublished PhD thesis, Univ Colorado, Boulder, 246 p
Deen JA, Rye RO, Munoz JL, Drexler JW (1994) The magmatic hydrothermal system at Julcani, Peru:
Evidence from fluid inclusions and hydrogen and oxygen isotopes. Econ Geol 89:1924-1938
Dole M, Lane GA, Rudd DP, Zaukelies DA (1954) Isotopic composition of atmospheric oxygen and
nitrogen. Geochim Cosmochim Acta 6:65-78
Eastoe CJ (1983) Sulfur isotope data and the nature of the hydrothermal systems at the Panguna and Frieda
porphyry copper deposits, Papua New Guinea. Econ Geol 78:201-213
Eastoe CJ, Gustin MM (1996) Volcanogenic massive sulfide deposits and anoxia in the Phanerozoic
oceans. Ore Geol Rev 10:179-197
Ebert SW, Rye RO (1997) Secondary precious metal enrichment by steam-heated fluids in the Crofoot-
Lewis hot spring gold-silver deposit and relation to paleoclimate. Econ Geol 92:578-600
Eckardt FD, Spiro B (1999) The origin of sulphur in gypsum and dissolved sulphate in the Central Namib
desert, Namibia. Sed Geol 123:255-273
596 Seal, Alpers & Rye
Elsgaard L, Jørgensen BB (1992) Anoxic transformations of radiolabeled hydrogen sulfide in marine and
freshwater sediments. Geochim Cosmochim Acta 56:2425-2435
Epstein S, Mayeda T (1953) Variations in the O18 content of waters from natural sources. Geochim
Cosmochim Acta 4:213-224
Field CW (1966) Sulfur isotopic method for discriminating between sulfates of hypogene and supergene
origin. Econ Geol 61:1428-1435
Field CW, Gustafson LB (1976) Sulfur isotopes in the porphyry copper deposit at EI Salvador, Chile. Econ
Geol 71:1533-1548
Fontes Je, Gonfiantini R (1967) Fractionnement isotopique de l’hydrogéne dans l’eau de cristallisation du
gypse. Acad Sci Comptes Rendus 265:4-6
Fournelle J (1990) Anhydrite in Nevado del Ruiz November pumice: Relevance to the sulfur problem. J
Volcanol Geoth Res 42:189-201
Friedman I (1953) Deuterium content of natural water and other substances. Geochim Cosmochim Acta
4:89-103
Fritz P, Basharmal GM, Drimmie RJ, Ibsen J, Qureshi RM (1989) Oxygen isotope exchange between
sulphate and water during bacterial reduction of sulphate. Chem Geol 79:99-105
Garlick GD (1969) The stable isotopes of oxygen. In Handbook of Geochemistry, 8B. Wedepohl KH (Ed)
Springer, Heidelberg, New York
Garrels RM,Thompson ME (1960) Oxidation of pyrite by iron sulfate solutions. Am J Sci 258A:57-67
Gavelin S, Parwel A, Ryhage R (1960) Sulfur isotope fractionation in sulfide mineralization. Econ Geol
55:510-530
Gemmell JB, Large RR (1992) Stringer system and alteration zones underlying the Hellyer volcanic-hosted
massive sulfide deposit, Tasmania, Australia. Econ Geol 87:620-649
Giesemann A, Jäger H-J, Norman AL, Krouse HR, Brand WA (1994) On-line sulfur-isotope determination
using an elemental analyzer coupled to a mass spectrometer. Anal Chem 66:2816-2819
Giggenbach W, Gonfiantini R, Panichi C (1983a) Geothermal systems. Guidebook on Nuclear Techniques
in Hydrology. International Atomic Energy Agency Tech Report Ser 91:359-379
Giggenbach WF, Gonfiantini R, Jangi BL, Truesdell AH (1983b) Isotopic and chemical composition of
Parbati Valley geothermal discharges, north-west Himalaya, India. Geothermics 12:199-222
Goldhaber MB (1983) Experimental study of metastable sulfur oxyanion formation during pyrite oxidation
at pH 6-9 and 30°C. Am J Sci 283:193-217
Goldhaber MB, Kaplan IR (1975) Controls and consequences of sulfate reduction rates in recent marine
sediments. Soil Sci 119:42-55
Gonfiantini R, Fontes JC (1963) Oxygen isotopic fractionation in the water of crystallization of gypsum.
Nature 200:644-646
Goodfellow WD, Jonasson IR (1984) Ocean stagnation and ventilation defined by δ34S secular trends in
pyrite and barite, Selwyn Basin, Yukon. Geology 12:583-586
Goodfellow WD, Franklin JM (1993) Geology, mineralogy, and chemistry of sediment-hosted clastic
massive sulfides in shallow cores, Middle Valley, northern Juan de Fuca Ridge. Econ Geol 88:2037-
2068
Green GR, Solomon M, Walshe JL (1981) The formation of the volcanic-hosted massive sulfide ore
deposit at Rosebery, Tasmania. Econ Geol 76:304-338
Hageman R, Nief G, Roth E (1970) Absolute isotopic scale for deuterium analysis of natural waters.
Absolute D/H ratio for SMOW. Tellus 22:712-715
Hamlin SN, Alpers CN (1996) Hydrogeology and geochemistry of acid mine drainage in ground water in
the vicinity of Penn Mine and Camanche Reservoir, Calaveras County, California: Second-year
summary. US Geol Survey Water-Resources Invest Report 96-4257, 44 P
Haur A, Hladikova J, Smejkal V (1973) Procedure of direct conversion of sulfates into SO2 for mass
spectrometric analysis of sulfur. Isotopenpraxis 18:433-436
Hayes JM, Lambert IB, Strauss H (1992) The sulfur-isotopic record. In The Proterozoic Biosphere. Schopf
JW, C Klein C (Eds) Cambridge Univ Press, 129-132
Heinzinger K (1969) Ein wasserstoffisotopieeffekt im CuSO4·5H2O. Z Naturforschung 24a:1502-1511
Hoering TC, Kennedy JW (1957) The exchange of oxygen between sulfuric acid and water. J Am Chem
Soc 79:56-60
Holland HD (1965) Some applications of thermochemical data to problems of ore deposits, II. Mineral
assemblages and the composition of ore-forming fluids. Econ Geol 60:1101-1166
Holland HD (1978) The Chemistry of the atmosphere and oceans. John Wiley and Sons, New York, 351 p
Holser WT, Kaplan IR (1966) Isotope geochemistry of sedimentary sulfates. Chem Geol 1:93-135
Holser WT, Kaplan IR, Sakai H, Zak I (1979) Isotope geochemistry of oxygen in the sedimentary sulfate
cycle. Chem Geol 25: 1-17
 Stable Isotope Systematics of Sulfate Minerals 597
Holt BD, Engelkemeier AG (1970) Thermal decomposition of barium sulfate to sulfur dioxide for mass
spectrometric analysis. Anal Chem 42:1451-1453
Holt BD, Kumar R, Cunningham PT (1981) Oxygen-18 study of the aqueous-phase oxidation of sulfur
dioxide. Atmos Environ 15:557-566
Horita J, Wesolowski DJ, Cole DR (1993) The activity-composition relationship of oxygen and hydrogen
isotopes in aqueous salt solutions: I. Vapor-liquid water equilibration of single salt solutions from 50
to 100°C. Geochim Cosmochim Acta 57:2797-2817
Horita J, Cole DR, Wesolowski DJ (1995) The activity-composition relationship of oxygen and hydrogen
isotopes in aqueous salt solutions: III. Vapor-liquid water equilibration of NaCI solutions to 350°C.
Geochim Cosmochim Acta 59:1139-1151
Hulston JR, Thode HG (1965a) Variations in the S33, S34, and S36 contents of meteorites and their relation
to chemical and nuclear effects. J Geophys Res 70:3475-3484
Hulston JR, Thode HG (1965b) Cosmic-ray produced S33 and S36 in the metallic phase of iron meteorites. J
Geophys Res 70:4435-4442
Huston DL (1999) Stable isotopes and their significance for understanding the genesis of volcanic-associated
massive sulfide deposits: A review In Volcanic-associated Massive Sulfide Deposits: Processes and
examples in modern and ancient settings. Barrie CT, Hannington (Eds) Rev Econc Geol 8:157-179
Igumnov SA, Grinenko VA, Poner NB (1977) Temperature dependence of the distribution coefficient of
sulfur isotopes between H2S and dissolved sulfates in the temperature range 260-400°C. Geokhimiya
7:1085-1087
Janecky DR, Shanks WC III (1988) Computational modeling of chemical and sulfur isotopic reaction
processes in seafloor hydrothermal systems: chimneys, massive sulfides, and subjacent alteration
zones. Can Mineral 26:805-825
Jones HD, Kesler SE, Furman FC, Kyle JR (1996) Sulfur isotope geochemistry of southern Appalachian
Mississippi Valley-type deposits. Econ Geol 91:355-367
Jørgensen BB, Isaksen MF, Jannasch HW (1992) Bacterial sulfate reduction above 100°C in deep-sea
hydrothermal vent sediments. Science 258:1756-1757
Kaiser CJ, Kelly WC, Wagner RJ, Shanks WC III (1987) Geologic and geochemical controls on
mineralization in the Southeast Missouri Barite district. Econ Geol 82:719-734
Kaplan IR, Hulston JR (1966) The isotopic abundance and content of sulfur in meteorites. Geochim
Cosmochim Acta 30:479-496
Kesler SE (1996) Appalachian Mississippi Valley-type deposits: Paleoaquifers and brine provinces. Soc
Econ Geol Spec Pub 4:29-57
Kesler SE, Jones HD, Furman FC, Sassen R, Anderson WH, Kyle JR (1994) Role of crude oil in the
genesis of Mississippi Valley-type deposits: Evidence from the Cincinnati Arch. Geology 22:609-612
Khademi H, Mermut AR, Krouse HR (1997) Isotopic composition of gypsum hydration water in selected
landforms from central Iran. Chem Geol 138:245-255
Kiyosu Y (1980) Chemical reduction and sulfur-isotope effects of sulfate by organic matter under
hydrothermal conditions. Chem Geol 30:47-56
Kornexl BE, Gehre M, Höfling R, Werner RA (1999) On-line δ18O measurement of organic and inorganic
substances. Rapid Commun Mass Spectrom 13:1685-1693
Kowalik J, Rye RO, Sawkins FJ (1981) Stable-isotope study of the Buchans, Newfoundland, polymetallic
sulphide deposit. Geol Assoc Canada Spec Paper 22:229-254
Kroopnick P, Craig H (1972) Atmospheric oxygen: isotopic composition and solubility fractionation.
Science 175:54-55
Krouse HR (1980) Sulphur isotopes in our environment. In Handbook of Environmental Isotope Geo-
chemistry I. The terrestrial environment. Fritz P, Fontes J-Ch (Eds.), Elsevier, Amsterdam, p 435-472
Krouse HR, Ueda A, Campbell FA (1990) Sulphur isotope abundances in coexisting sulphate and sulphide:
Kinetic isotope effects versus exchange phenomena. In Stable Isotopes and Fluid Processes in
Mineralization. Herbert HK, Ho SE (Eds) Univ of Western Australia, Univ Extension Pub 23:226-243
Kusakabe M (1974) Sulphur isotopic variations in nature, 10. Oxygen and sulphur isotope study of
Wairakei (New Zealand) geothermal well discharges. New Zealand J Sci 17:183-191
- -
Kusakabe M, Robinson BW (1977) Oxygen and sulfur isotope equilibria in the BaSO4 HSO4 -H2O system
from 110 to 350°C with applications. Geochim Cosmochim Acta 41:1033-11040
Kusakabe M, Chiba H (1983) Oxygen and sulfur isotope composition of barite and anhydrite from the
Fukazawa deposit, Japan. Econ Geol Monogr 5:292-301
Lambert IB, Donnelly TH (1990) The paleoenvironmental significance trends in sulphur isotope
compositions in the Precambrian: A critical review. In Stable Isotopes and Fluid Processes in
Mineralization. Herbert HK, Ho SE (Eds) Univ of Western Australia, Univ Extension Pub 23:260-268
Lawrence JR, Taylor HP Jr. (1971) Deuterium and oxygen-18 correlation: clay minerals and hydroxides in
Quaternary soils compared to meteoric waters. Geochim Cosmochim Acta 35:993-1004
598 Seal, Alpers & Rye
Leggett JK (1980) British Lower Paleozoic black shales and their palaeo-oceanographic significance. J
Geol Soc London 137:139-156
Leuth VW, Rye RO, Peters L (2000) Hydrothermal “sour gas” jarosite: Ancient and modern acid sulfate
mineralization events in the Rio Grande rift. Geol Soc Am Abstracts with Programs (in press)
Lloyd RM (1967) Oxygen-18 composition of oceanic sulfate. Science 156:1228-1231
Lloyd RM (1968) Oxygen isotope behavior in the sulfate-water system. J Geophys Res 73:6099-6110
Longinelli A, Craig H (1967) Oxygen-18 variations in sulfate ions in sea water and saline lakes. Science
156:56-59
Luhr JF (1990) Experimental phase relations of water-and sulfur-saturated arc magmas and the 1982
eruptions of El Chichón volcano. J Petrol 31:1071-1114
Luhr JF, Carmichael ISE, Varenkamp JC (1984) The 1982 eruptions of El Chichón volcano, Chiapas,
Mexico: Mineralogy and petrology of the anhydrite-bearing pumices. J Volcanol Geotherm Res
23:69108
Macnamara J, Thode HG (1950) Comparison of the isotopic constitution of terrestrial and meteoritic
sulphur. Phys Rev 78:307
Martinez-S. RG, Jacquier B, Arnold M (1996) The δ34S composition of sulfates and sulfides at the Los
Humeros geothermal system, Mexico and their application to physicochemical fluid evolution. J
Volcanol Geotherm Res 73:99-118
Matsuo S, Kusakabe M, Niwano M, Hirano T, Oki Y (1985) Origin of thermal waters from the Hakone
geothermal system, Japan. Geochem J 19:27-44
Maynard JB, Okita PM (1991) Bedded barite deposits in the United States, Canada, Germany, and China:
two major types based on tectonic setting. Econ Geol 86:364-376
McGoldrick PJ, Large RR (1992) Geologic and geochemical controls on gold-rich stringer mineralization
in the Que River deposit, Tasmania. Econ Geol 87:667-685
McKenzie WF, Truesdell AH (1977) Geothermal reservoir temperatures estimated from the oxygen isotope
compositions of dissolved sulfate and water from hot springs and shallow drillholes. Geothermics
5:51-61
McKibben MA, Barnes HL (1986) Oxidation of pyrite in low temperature acidic solutions: rate laws and
surface textures. Geochim Cosmochim Acta 50:1509-1520
McKibben MA, Eldridge CS, Reyes AG (1996) Sulfur isotopic systematics of the June 1991 Mount
Pinatubo eruptions: A SHRIMP ion microprobe study. In Fire and mud⎯eruptions and lahars of
Mount Pinatubo, Philippines. Newhall CG, Punongbayan RS (Eds) PHIVOLCS and Univ
Washington Press, 825-843
McKibben MA, Riciputi LR (1998) Sulfur isotopes by ion microprobe. In Applications of microanalytical
techniques to understanding mineralizing processes. McKibben MA, Shanks WC III, Ridley WI (Eds)
Reviews in Economic Geology 7:121-139
Misra KC, Gratz JF, Lu C (1996) Carbonate-hosted Mississippi Valley-type mineralization in the
Elmwood-Gordonsville deposits, Central Tennessee zinc district: A synthesis. Soc Econ Geol Spec
Pub 4:58-73
Mizutani Y (1972) Isotopic composition and underground temperature of the Otake geothermal water,
Kyushu, Japan. Geochem J 6:67-73
Mizutani Y, Rafter TA (1969a) Oxygen isotopic composition of sulphates, Part 3: Oxygen isotopic
fractionation in the bisulphate ion-water system. New Zealand J Sci 12:54-59
Mizutani Y, Rafter TA (1969b) Oxygen isotopic composition of sulphates, Part 4: Bacterial fractionation of
oxygen isotopes in the reduction of sulphate and in the oxidation of sulphur. New Zealand J Sci
12:60-68
Mizutani Y, Rafter TA (1973) Isotopic behavior of sulphate oxygen in the bacterial reduction of sulphate.
Geochem J 6:183-191
Moses CO, Nordstrom DK, Herman JS, Mills AL (1987) aqueous pyrite oxidation by dissolved oxygen and
ferric iron. Geochim Cosmochim Acta 51:1561-1571
Nakai N, Jensen (1964) The kinetic isotope effect in the bacterial reduction and oxidation of sulfur.
Geochim Cosmochim Acta 28:1893-1912
Nielsen H, Pilot J, Grinenko LN, Grinenko VA, Lein AY, Smith JW, Pankina RG (1991) Lithospheric
sources of sulphur. In Stable Isotopes in the Assessment of Natural and Anthropogenic Sulphur in the
Environment. Krouse HR, Grinenko VA (Eds) SCOPE 43, J Wiley and Sons, New York, 65-132
Nordstrom DK, Alpers CN (1999a) Geochemistry of acid mine waters. In The Environmental
Geochemistry of Mineral Deposits, Part A: Processes, techniques, and health issues. Plumlee GS,
Logsdon MJ (Eds) Rev Econ Geol 6A:133-160
Nordstrom DK, Alpers CN (1999b) Negative pH, efflorescent mineralogy, and consequences for
environmental restoration at the Iron Mountain Superfund site, California. Proc Nat Acad Sci
96:3455-3462
 Stable Isotope Systematics of Sulfate Minerals 599
Nordstrom DK, Alpers CN, Ptacek CJ, Blowes DW (2000) Negative pH and extremely acidic mine waters
from Iron Mountain, California. Environ Sci Techn 34:254-258
Nriagu JO, Ress CE, Mekhtiyeva VL, Lein AY, Fritz P, Drimmie RJ, Pankina RG, Robinson RW, Krouse
HR (1991) Hydrosphere. In Stable Isotopes in the Assessment of Natural and Sulphur in the
Environment. Krouse HR, Grinenko VA (Eds) SCOPE 43, John Wiley and Sons, New York, p 177-
265
Nuelle LM, Shelton KL (1986) Geologic and geochemical evidence of possible bedded barite deposits in
Devonian rocks of the Valley and Ridge Province, Appalachian Mountains. Econ Geol 81: 1408-1430
Ohmoto, H (1972) Systematics of sulfur and carbon isotopes in hydrothermal ore deposits. Econ Geol
67:551-578
Ohmoto, H (1986) Stable isotope geochemistry of ore deposits. Rev Mineral 16:185-225
Ohmoto H (1992) Biogeochemistry of sulfur and the mechanisms of sulfide-sulfate mineralization in
Archean oceans. In Early Organic Evolution: Implications for Mineral and Energy Resources.
Schidlowski Met al. (Eds) Springer-Verlag, p 378-397
Ohmoto H, Rye RO (1974) Hydrogen and oxygen isotopic compositions of fluid inclusions in the Kuroko
deposits, Japan. Econ Geol 69:947-953
Ohmoto H, Rye RO (1979) Isotopes of sulfur and carbon. In Geochemistry of hydrothermal ore deposits.
Barnes HL (Ed) J Wiley and Sons, New York, p 509-567
Ohmoto H, Lasaga AC (1982) Kinetics of reactions between aqueous sulfates and sulfides in hydrothermal
systems. Geochim Cosmochim Acta 46:1727-1745
Ohmoto H, Felder RP (1987) Bacterial activity in the warmer, sulphate-bearing, Archaean oceans. Nature
328:244-246
Ohmoto H, Goldhaber (1997) Sulfur and carbon isotopes. In Geochemistry of Hydrothermal Ore Deposits.
Barnes HL (Ed) J Wiley and Sons, New York, p 517-611
O’Neil JR (1986) Theoretical and experimental aspects of isotopic fractionation. Rev Mineral 16:1-40
O’Neil JR, Clayton RN, Mayeda TK (1969) Oxygen isotope fractionation in divalent metal carbonates. J
Phys Chem 51:5547-5558
O’Neil JR, Silberman ML (1974) Stable isotope relations in epithermal Au-Ag deposits. Econ Geol 69:902-
909
Orr WL (1974) Changes in sulfur content and isotopic ratios of sulfur during petroleum maturation⎯study
of Big Horn Paleozoic oils. Am Assoc Petrol Geol Bull 58:2295-2318
Oudin E, Thisse Y, Ramboz C (1984) Fluid inclusion and mineralogical evidence for high-temperature
saline hydrothermal circulation in the Red Sea metalliferous sediments: preliminary results. Mar
Mining 5:3-31
Paytan A, Kastner M, Campbell D, Thiemens MH (1998) Sulfur isotopic composition of Cenozoic
seawater sulfate. Science 282:1459-1462
Peter JM, Shanks WC III (1992) Sulfur, carbon, and oxygen isotope variations in submarine hydrothermal
deposits of the Guaymas Basin, Gulf of California, USA. Geochim Cosmochim Acta 56:2025-2040
Pickthorn WJ, O’Neil JR (1985) 18O relations in alunite minerals: potential single-mineral geothermometer.
Geol Soc Am Abstracts with Program 17:689
Plummer LN, Parkhurst DL, Fleming GW, Dunkle SA (1988) A computer program incorporating Pitzer’s
equations for calculation of geochemical reactions in brines. US Geol Survey Water-Resources Invest
Report 88-4153
Pottorf RJ, Barnes HL (1983) Mineralogy, geochemistry and ore genesis of hydrothermal sediments from
the Atlantis II Deep, Red Sea. Econ Geol Monogr 5:198-223
Postgate JR (1984) The Sulfate-Reducing Bacteria. 2nd Ed., Cambridge University Press, Cambridge, UK
Rafter TA (1967) Oxygen isotopic compositions of sulphates, Part 1: A method for the extraction of
oxygen and its quantitative conversion to carbon dioxide for isotopic radiation methods. New Zealand
J Sci 10:493-515
Rees CE, Holt BD (1991) The isotopic analysis of sulphur and oxygen. In Stable Isotopes in the
Assessment of Natural and Anthropogenic Sulphur in the Environment. Krouse HR, Grinenko VA
(Eds) SCOPE 43, J Wiley and Sons, New York, p 43-64
Rees CE, Jenkins WJ, Monster J (1978) The sulphur isotope geochemistry of ocean water sulphate.
Geochim Cosmochim Acta 42:377-382
Richardson CK, Rye RO, Wasserman MD (1988) The chemical and thermal evolution of the fluids in the
Cave-in-Rock fluorspar district, Illinois: stable isotope systematics at the Deardorff mine. Econ Geol
83:765-783
Rickard DT, Zweifel H, Donnelly TH (1979) Sulfur isotope systematics in the Åsen pyrite-barite deposits,
Skellefte district, Sweden. Econ Geol 74:1060-1068
Rimstidt JD (1997) Gangue mineral transport and deposition. In Geochemistry of Hydrothermal Ore
Deposits. Barnes HL (Ed) J Wiley and Sons, New York, p 487-515
600 Seal, Alpers & Rye
Robinson BW (1973) Sulfur isotope equilibrium during sulfur hydrolysis at high temperatures. Earth Planet
Sci Lett 18:443-450
Robinson BW (1978) Sulphate-water and -H2S isotopic thermometry in the New Zealand geothermal
systems. US Geol Survey Open-file Report 78-701:354-356
Ross GM, Bloch JD, Krouse HR (1995) Neoproterozoic strata of the southern Canadian Cordillera and the
isotopic evolution of seawater sulfate. Precambrian Res 73:71-99
Rouchy JM, Camoin G, Casanova J, Deconinck JF (1993) The central palaeo-Andean basin of Bolivia
(Potosi area) during the late Cretaceous and early Tertiary: Reconstruction of ancient saline lakes
using sedimentological, palaeoecological· and stable isotope records. Palaeogeogr Palaeoclim
Palaeoecol 105:179-198
Rozanski K, Araguás- Araguás L, Gonfiantini R (1993) Isotopic patterns in modern global precipitation. In
Climate Change in Continental Isotopic Records. Swart PK, Lohmann KC, McKenzie JA, Savin S
(Eds) Geophys Monogr 78:1-36
Rye RO (1966) The carbon, hydrogen, and oxygen isotopic composition of the hydrothermal fluids
responsible for the lead-zinc deposits at Providencia, Zacatecas, Mexico. Econ Geol61:1399-1427
Rye RO (1993) The evolution of magmatic fluids in the epithermal environment: the stable isotope
perspective. Econ Geol 88:733-753
Rye RO, Alpers CN (1997) The stable isotope geochemistry of jarosite. US Geol Survey Open-file Report
97-88, 28 p
Rye RO, Ohmoto H (1974) Sulfur and carbon isotopes and ore genesis: A review. Econ Geol 69:826-842
Rye RO, Stoffregen RE (1995) Jarosite-water oxygen and hydrogen isotope fractionations: preliminary
experimental data. Econ Geol 90:2336-2342
Rye RO, Stoffregen RE, Bethke PM (1990) Stable isotope systematics and magmatic hydrothermal
processes in the Summitville, CO, gold deposit. US Geol Survey Open-File Report 90-626, 31 p
Rye RO, Shawe DR, Poole FG (1978) Stable isotope studies of bedded barite at East Northumberland
Canyon in Toquima Range, central Nevada. US Geol Survey J Res 6:221-229.
Rye RO, Luhr JF, Wasserman MD (1984) Sulfur and oxygen isotope systematics of the 1982 eruptions of
El Chichón volcano, Chiapas, Mexico. J Volcanol Geotherm Res 23:109-123
Rye RO, Plumlee GS, Barton PM, Barton PB (1988) Stable isotope geochemistry of the Creede, Colorado,
hydrothermal system. US Geological Survey Open-File Report 88-356, 40 p
Rye RO, Bethke PM, Wasserman MD (1992) The stable isotope geochemistry of acid sulfate alteration.
Econ Geol 87:225-262
Rye RO, Bethke PM, Finkelstein DB (2000a) Stable isotope evolution and paleolimnology of Ancient Lake
Creede. In Ancient Lake Creede, its volcano-tectonic setting, sedimentation history, and relation to
mineralization in the Creede mining district. Bethke PM, Hay RL (Eds) Geol Soc Am Spec Paper
346:233-266
Rye RO, Bethke PM, Lanphere MA, Steven TA (2000b) Neogene geomorphic and climatic evolution of
the central San Juan Mountains, CO: K/Ar age and stable isotope data on supergene alunite and
jarosite from the Creede mining district. In Ancient Lake Creede, its volcano-tectonic setting,
sedimentation history, and relation to mineralization in the Creede mining district. Bethke PM, Hay
RL (Eds) Geol Soc Am Spec Paper 346:95-104
Sakai H (1968) Isotopic properties of sulfur compounds in hydrothermal processes. Geochem J 2:29-49
Sakai H (1977) Sulfate-water isotope thermometry applied to geothermal systems. Geothermics 5:67-74
Sakai H (1983) Sulfur isotope exchange rate between sulfate and sulfide and its application. Geothermics
12:111-117
Sakai H, Krouse HR (1971) Elimination of memory effects in 18O/16O determinations in sulphates. Earth
Planet Sci Lett 11:369-373
Sakai H, Matsubaya O (1977) Stable isotopic studies of Japanese geothermal systems. Geothermics 5:97-
124
Sakai H, Dickson FW (1978) Experimental determination of the rate and equilibrium fractionation factors
of sulfur isotope exchange between sulfate and sulfide in slightly acid solutions at 300°C and 1000
bars. Earth Planet Sci Lett 39:151-161
Sakai H, Takenaka T, Kishima N (1980) Experimental study of the rate and isotope effect in sulfate
reduction by ferrous oxides and silicates under hydrothermal conditions. Proc 3rd Int’l Symp Water-
Rock Interact, Edmonton, Alberta, p 75-76
Sakai H, Des Marais DJ, Ueda A, Moore JG (1984) Concentrations and isotope ratios of carbon, nitrogen,
and sulfur in ocean-floor basalts. Geochim Cosmochim Acta 48:2433-2442
Schwarcz HP, Harmon RS, Thompson, P, Ford DC (1976) Stable isotope studies of fluid inclusions in
speleothems and their paleoclimatic significance. Geochim Cosmochim Acta 40:657-665
Seal RR II, Rye RO (1992) Stable isotope study of water-rock interaction and ore formation, Bayhorse base
and precious metal district, Idaho. Econ Geol 87:271-287
 Stable Isotope Systematics of Sulfate Minerals 601
Seal RR II, Rye RO (1993) Stable isotope study of fluid inclusions in fluorite from Idaho: implications for
continental climates during the Eocene. Geology 21:219-222
Seal RR II, Wandless GA (1997) Stable isotope characteristics of waters draining massive sulfide deposits
in the eastern United States. In Fourth Int’l Symp Environ Geochem Proc. Wanty RB, Marsh SP,
Gough LP (Eds) US Geol Survey Open-File Report 97-496, 82 p
Seal RR II, Ayuso RA, Foley NK, Clark SHB (2000) Sulfur and lead isotope geochemistry of hypogene
mineralization at the Barite Hill gold deposit, Carolina Slate Belt, southeastern United States: a
window into and through regional metamorphism. Mineral Deposita (in press)
Seccombe PK, Godwin CI, Krouse HR, Juras SJ (1990) Sulphur and lead isotopic studies of the Buttle
Lake massive sulphide deposits, Vancouver Island, B.C.: Sources of ore constituents in a submarine
exhalative environment. Proc Pacific Rim 90 Congress, Gold Coast, p 419-426
Seyfried WE, Bischoff JL (1981) Experimental seawater-basalt interaction at 300°C, 500 bars, chemical
exchange, secondary mineral formation and implications for the transport of heavy metals. Geochim
Cosmochim Acta 45:135-149
Shanks WC III, Bischoff JL (1980) Geochemistry, sulfur isotope composition, and accumulation rates of
the Red Sea geothermal deposits. Econ Geol 75:445-459
Shanks WC III, Seyfried WE Jr. (1987) Stable isotope studies of vent fluids and chimney minerals,
southern Juan de Fuca Ridge: sodium metasomatism and seawater sulfate reduction. J Geophys Res
92:1387-11399
Shanks WC III, Bischoff JL, Rosenbauer RJ (1981) Seawater sulfate reduction and sulfur isotope
fractionation in basaltic systems: interaction of seawater with fayalite and magnetite at 200-350°C.
Geochim Cosmochim Acta 45:1977-1995
Shanks WC III, Böhlke JK, Seal RR II (1995) Stable isotopes in mid-ocean ridge hydrothermal systems:
interactions between fluids, minerals, and organisms. In Seafloor hydrothermal systems: Physical,
chemical, biological, and geological interactions. Humphris SE, Zierenberg RA, Mullineaux LS,
Thomson RE (Eds) Geophys Monogr 91:194-221
Shelton KL, Rye DM (1982) Sulfur isotopic compositions of ores from Mines Gaspé, Quebec: an example
of sulfate-sulfide isotopic disequilibria in ore-forming fluids with applications to other porphyry type
deposits. Econ Geol 77:1688-1709
Sheppard SNIF, Nielsen RL, Taylor HP Jr. (1969) Oxygen and hydrogen isotope ratios of clays from
porphyry copper deposits. Econ Geol 64:755-777
Simmons SF, Stewart MK, Robinson BW, Glover RB (1994) The chemical and isotopic compositions of
thermal waters at Waimangu, New Zealand. Geothermics 23:539-553
Singer PC, Stumm W (1970) Acid mine drainage: the rate-determining step. Science 167:1121-1123
Sofer Z, Gat JR (1975) The isotope composition of evaporating brines: effect of the isotopic activity ration
in saline solutions. Earth Planet Sci Lett 26:179-186
Solomon M, Rafter TA, Jensen ML (1969) Isotope studies on the Rosebery, Mount Farrell and Mount Lyell
ores, Tasmania. Mineral Deposita 4:172-199
Solomon M, Eastoe CJ, Walshe JL, Green GR (1988) Mineral deposits and sulfur isotope abundances in
the Mount Read Volcanics between Que River and Mount Darwin, Tasmania. Econ Geol 83:1307-
1328
Stewart MK (1974) Hydrogen and oxygen isotope fractionation during crystallization of mirabilite and ice.
Geochim Cosmochim Acta 38:167-172
Stoffregen RE (1993) Stability relations of jarosite and natrojarosite at 100-250°C. Geochim Cosmochim
Acta 57:2417-2429
Stoffregen RE, Rye RO, Wasserman MD (1994) Experimental studies of alunite: 1. 18O-16O and D-H
fractionation factors between alunite and water at 250-450°C. Geochim Cosmochim Acta 58:903-916
Strauss H (1993) The sulfur isotopic record of Precambrian sulfates: new data and a critical evaluation of
the existing record. Precambrian Res 63:25-246
Taylor BE (1986) Magmatic volatiles: isotopic variation of C, H, and S. Rev Mineral 16:185-225
Taylor BE, Wheeler MC (1994) Sulfur- and oxygen-isotope geochemistry of acid mine drainage in the
western United States. In Environmental Geochemistry of Sulfide Oxidation. Alpers CN, Blowes DW
(Eds) Am Chem Soc Symp Ser 550:481-514
Taylor BE, Wheeler MC, Nordstrom DK (1984) Stable isotope geochemistry of acid mine drainage:
experimental oxidation of pyrite. Geochim Cosmochim Acta 48:2669-2678
Taylor BE, O’Neil JR (1977) Stable isotope studies of metasomatic skarns and associated metamorphic and
igneous rocks, Osgood Mountains, Nevada. Contrib Mineral Petrol 63:1-49
Taylor HP Jr. (1997) Oxygen and hydrogen isotope relationships in hydrothermal mineral deposits. In
Geochemistry of Hydrothermal Ore Deposits. Barnes HL (Ed) J Wiley and Sons, New York, 229-302
Thiemens MH (1999a) Mass-independent isotope effects in planetary atmospherics and the early solar
system. Science 283:341-345
 602 Seal, Alpers & Rye
Thiemens MH (1999b) Mass independent isotopic composition of atmospheric molecules. Late abstract
added to 9th Goldschmidt Conf Abstract with Programs
Thode HG, Monster J (1965) Sulfur-isotope geochemistry of petroleum, evaporites, and ancient seas. Am
Assoc Petrol Geol Mem 4:367-377
Toran L, Harris RF (1989) Interpretation of sulfur and oxygen isotopes in biological and abiological sulfide
oxidation. Geochim Cosmochim Acta 53:2341-2348
Trudinger PA, Chambers LA (1973) Reversibility of bacterial sulfate reduction and its relevance to isotope
fractionation. Geochim Cosmochim Acta 37:1775-1778
Truesdell AH, Nathenson M, Rye RO (1977) The effects of subsurface boiling and dilution on the isotopic
compositions of Yellowstone thermal waters. J Geophys Res 82:3694-3704
Truesdell AH, Hulston JR (1980) Isotopic evidence of environments of geothermal systems. In Handbook
of Environmental Isotope Geochemistry 1:79-226. Fritz P, Fontes J-Ch (Eds)
Ueda A, Sakai H (1984) Sulfur isotope study of Quaternary volcanic rocks from the Japanese island arc.
Geochim Cosmochim Acta 44:579587
Urey HC (1947) The thermodynamic properties of isotopic substances. J Chem Soc 1947:562-581
Utrilla R, Pierre C, Orti F, Pueyo JJ (1992) Oxygen and sulphur isotope compositions as indicators of the
origin of Mesozoic and Cenozoic evaporites from Spain. Chem Geol 102:229-244
Van Everdingen RO, Krouse HR (1988) Interpretation of isotopic compositions of dissolved sulfates in
acid mine drainage. US Bur Mines Info Circ 9183:147-156
Van Stempvoort DR, Krouse HR (1994) Controls on δ18O in sulfate. In Environmental Geochemistry of
Sulfide Oxidation. Alpers CN, Blowes DW (Eds) Am Chem Soc Symp Ser 550:446-480
Veizer J, Bruckschen P, Pawellek F, Diener A, Podlaha OG, Carden GAF, Jasper T, Korte C, Strauss H,
Azmy K, Ala D (1997) Oxygen isotope evolution of Phanerozoic seawater. Palaeogeogr Palaeoclim
Palaeoecol 132:159-172
Vennemann TW, O’Neil JR (1993) A simple and inexpensive method of hydrogen isotope and water
analyses of minerals and rocks based on zinc reagent. Chem Geol 103:227-234
Wakshal E, Nielsen H (982) Variations of δ34S(SO4), δ18O(H2O) and Cl/SO4 ratio in rainwater over
northern Israel, from the Mediterranean coast to Jordan rift valley and Golan Heights. Earth Planet Sci
Lett 61:272-282
Wang Z, Li G (1991) Barite and witherite deposits in Lower Cambrian shales of south China: stratigraphic
distribution and geochemical characterization. Econ Geol 86:354-363
Wasserman MD, Rye RO, Bethke PM, Arribas A Jr (1992) Methods for separation and total stable isotope
analysis of alunite. US Geol Survey Open-File Report 92-9
Watanabe M, Sakai H (1983) Stable isotope geochemistry of sulfates from the Neogene ore deposits in the
Green Tuff region, Japan. Econ Geol Monogr 5:282-291
Way K, Fano L, Scott MR, Thew K (1950) Nuclear data. A collection of experimental values of halflifes,
radiation energies, relative isotopic abundances, nuclear moments and cross-sections. Nat Bur
Standards US Circ 499
Whelan JF, Rye RO, deLorraine W (1984) The Bahnat-Edwards zinc-lead deposits⎯synsedimentary ore
from Mississippi Valley-type fluids. Econ Geol 79:239-265
Whelan JF, Rye RO, deLorraine W, Ohmoto H (1990) Isotopic geochemistry of a mid-Proterozoic
evaporite basin: Balmat, New York. Am J Sci 290:396-424
Wiersma CL, Rimstidt JD (1984) Rates of reaction of pyrite and marcasite with ferric iron at pH 2.
Geochim Cosmochim Acta 48:85-92
Yurtsever Y, Gat JR (1981) Atmospheric waters. In1’l Atomic Tech Report 210:103-142
Zak I, Sakai H, Kaplan IR (1980) Factors controlling 18O/16O and 34S/32S isotope ratios of ocean sulfates,
evaporites and interstitial sulfates from modern deep sea sediments. Isotope Marine Chem 339-373
Zierenberg RA, Shanks WC III (1986) Isotopic constraints on the origin of the Atlantis II, Suakin and
Valdivia brines, Red Sea. Geochim Cosmochim Acta 50:2205-2214
Zierenberg RA, Shanks WC ITI (1988) Isotopic studies of epigenetic features in metalliferous sediment,
Atlantis II Deep, Red Sea. Can Mineral 26:737-753
Zierenberg RA, Shanks WC III (1994) Sediment alteration associated with massive sulfide formation in the
Escanaba Trough, Gorda Ridge: the importance of seawater mixing and magnesium metasomatism.
US Geol Survey Bull 2022:257-277
Zierenberg RA, Koski RA, Morton IL, Bouse RM, Shanks WC ITI (1993) Genesis of massive sulfide
deposits on a sediment-covered spreading center, Escanaba Trough, southern Gorda Ridge. Econ Geol
88:2069-2098

View publication stats