Journal of Petrology Advance Access originally published online on January 7, 2005
Journal of Petrology 2005 46(4):783-798; doi:10.1093/petrology/egh097
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
An Experimental Study of the Sulfur Content in Basaltic Melts Saturated with Immiscible Sulfide or Sulfate Liquids at 1300°C and 1·0 GPa
DEPARTMENT OF EARTH & ATMOSPHERIC SCIENCES, UNIVERSITY OF ALBERTA, EDMONTON, ALBERTA, CANADA, T6G 2E3
RECEIVED JANUARY 10, 2004; ACCEPTED NOVEMBER 18, 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The sulfur content in basaltic melts coexisting with either sulfide or sulfate melts was determined experimentally. The experimental conditions were in the range of 1300–1355°C and 1·0–1·6 GPa, conditions appropriate for the melting of the upper mantle above subduction zones. Under these conditions, both sulfide and sulfate were present as immiscible liquids, as inferred from the round geometries of the quenched sulfide and sulfate phases. The measured S content in basaltic melts saturated with sulfate liquids ([S] = 1·5 ± 0·2 wt %) was 10 times higher than the S content in basaltic melts saturated with sulfide liquids ([S] = 0·14 ± 0·02 wt %). In our experiments, sulfate liquids were stable at fO2 as low as
FMQ = +1·85 [FMQ = log (fO2)sample – log (fO2)FMQ, where FMQ is the fayalite–magnetite–quartz oxygen buffer], and evidence from other sources indicates that sulfates will be stable at lower fO2 in melts with lower activities of silica. Because chalcophile and highly siderophile elements, such as Cu, Ni, Au, and Pd, are partitioned preferentially into sulfide phases, melting of sufficiently oxidized sources, in which sulfides are not stable, would favor incorporation of these elements into the silicate melt produced. Such melts would have a higher potential to generate ore deposits. This study shows that the high sulfur contents of such oxidized basalts also means that relatively small amounts of such magmas can provide significant amounts of sulfur to exsolving volatile phases and account for the bulk of the sulfur expelled in some volcanic eruptions, such the 1991 eruption of Mount Pinatubo. KEY WORDS: basalt; mantle; oxidation state; sulfate; sulfur
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Although sulfur is a minor element in the Earth (250 ± 50 ppm, the 11th most abundant element in the Silicate Earth; McDonough & Sun, 1995
), it has a disproportionate impact on the geochemical behavior of some elements because S commonly occurs as discrete sulfide phases. Thus, sulfides largely control the behavior of chalcophile (e.g. Cu, Ni) and highly siderophile elements (Ru, Rh, Pd, Re, Os, Ir, Pt, and Au), elements that are of interest because they provide valuable information about geochemical processes and because they are economically important. For these reasons, the solubility of S as sulfide and the stability of sulfide minerals in silicate melts have been widely investigated experimentally as functions of pressure, temperature, and melt composition (e.g. see Poulson & Ohmoto, 1990, for a compilation of experimental work until 1990; Mavrogenes & O'Neill, 1999; Holzheid & Grove, 2002; O'Neill & Mavrogenes, 2002). Consequently, these data have been used to model sulfide elimination in magmatic sources and sulfide precipitation in crystallizing melts.
Although occurrences of sulfate-bearing minerals in igneous rocks (e.g. nosean, haüyne, silvialite) were documented early in the 20th century (e.g. Lindgren & Ransome, 1906; Brauns, 1914), the assumption that S is present only as sulfide in magmatic systems was challenged only by the discovery of anhydrite in pumices erupted by El Chichón (Chiapas, Mexico) in 1982 (Luhr et al., 1984). Luhr & Varekamp (1984) compared the anhydrite-bearing eruptions of El Chichón in early 1982 to the eruption of carbonatite lavas from Oldoinyo Lengai in 1960 that established the existence of carbonatitic melts. They recognized the implications of igneous anhydrite occurrences and concluded that the El Chichón eruptions ‘seem destined to alter many of our concepts regarding the role of S in magmatism and the relationship between explosive volcanism and the earth's climate’ (Luhr & Varekamp, 1984, p. vii).
Subsequent experimental studies (Carroll & Rutherford, 1985, 1987; Luhr, 1990) showed that sulfates (e.g. anhydrite), not sulfides, are the stable S-bearing magmatic phases at oxygen fugacities (fO2) above those defined by the nickel–nickel oxide (bunsenite) oxygen buffer (NNO). [FMQ is used in this paper to indicate relative oxygen fugacity, where FMQ = log (fO2)sample – log (fO2)FMQ, and FMQ refers to the fayalite–magnetite–quartz oxygen buffer; for comparison purposes NNO 
FMQ – 0·7, although the exact relationship is a function of P and T.] Estimates of the oxidation state of island-arc basalts (IAB) show that they are typically more oxidized than basalts from other tectonic settings (Ballhaus, 1993; Parkinson & Arculus, 1999) and estimates of the speciation of S in some basaltic systems show a dominance of sulfate over sulfide (Nilsson & Peach, 1993; Metrich & Clocchiatti, 1996; Gurenko & Schminke, 1998, 2000; Matthews et al., 1999). Therefore, solubility data for S as sulfate are needed to provide appropriate models for the behavior of S, and of chalcophile and highly siderophile elements in these environments.
Eruption products from the 1991 eruption of Mt. Pinatubo (Philippines) showed once more that anhydrite was a stable igneous phase. This eruption released more than 17 Mt of SO2 into the atmosphere (Gerlach et al., 1996) and demonstrated the significant short-term impact that high-S explosive volcanism can have on global climate (e.g. cooling of the order of 0·5°C over large parts of the Earth and significant increase of the ozone depletion rates; Self et al., 1996). At Mt. Pinatubo, intrusion of basaltic magmas into a large gas-charged dacitic body followed by explosive eruption is well documented for the 1991 eruption and for prehistoric eruptions (Newhall et al., 1996; Pallister et al., 1996). De Hoog et al. (2003) estimated that the oxidation state of the basaltic melt involved in this eruption was similar to that of the dacitic magma (FMQ = +2·1), showing that mixing between the dacitic and the basaltic magma did not result in significant changes in fO2 as suggested in some models for the 1991 Pinatubo eruption (e.g. Hattori, 1993; Kress, 1997).
Estimates of the stratospheric loading of H2SO4 following the eruption of the Huaynaputina, Peru, in 1600 are consistent with this eruption causing the low temperatures for the summer of 1601, which were among the coldest in 1500 years in Fennoscandia (de Silva & Zielinski, 1998). A similar temporal link has been established between the low temperatures recorded in 1816 (‘the year without summer’) and the eruption of Tambora (Sumbawa island, Indonesia) in April 1815. This eruption produced an estimated stratospheric loading of 
200 Mt H2SO4 and is considered the greatest S-producing eruption of the last 750 years (Stothers, 1984; de Silva & Zielinski, 1998; Palmer et al., 2001). If high-S volcanic eruptions are associated with, and perhaps triggered by, oxidized basaltic melts (as in Mt. Pinatubo), modeling of these processes needs to account for S being present as sulfate in the basalts.
In this paper we present the first experimental data on the S content in basaltic melts saturated with sulfate phases, and show that partial melting of oxidized mantle sources will produce magmas enriched in sulfur and chalcophile and siderophile elements relative to those produced from sulfide-saturated sources. Such magmas may have elevated potential for forming ore deposits of these elements, and may contribute to the S budget in explosive volcanism.
|
EXPERIMENTAL AND ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Starting materials
Three sets of experiments were performed in this study: (1) sulfide-saturated runs, produced from sulfide-bearing starting materials; (2) sulfate-saturated runs, produced from sulfate-bearing starting materials; (3) sulfide-saturated runs, produced by reduction from sulfate-bearing starting materials (‘sulfate-reduction runs’). Each set of experiments required the combination of different starting materials; however, bulk compositions and experimental conditions were kept as similar as possible to isolate the effect of oxidation state on the behavior of S.
Starting materials that would yield a basaltic melt coexisting with a lherzolitic residue were used to simulate conditions of partial melting under pressures and temperatures appropriate for the upper mantle. Because these experiments were also used to study the behavior of Pd, Ir, Pt, and Au, small amounts of these metals were also added. The results pertaining to the behavior of these metals will be presented elsewhere.
Glasses
Three synthetic glasses (compositions listed in Table 1) were prepared as starting material. Suitable chemical compositions were selected among various melt compositions from previous experimental work on partial melting of peridotitic mantle (e.g. Jaques & Green, 1980; Gaetani & Grove, 1998; Robinson et al., 1998). We used chemical compositions of starting materials and run products from the experimental work of Gaetani & Grove (1998), simplified by eliminating Mn, K, and P, which are minor or trace elements that are unlikely to affect the phase equilibria studied here. One glass (labeled B302) was prepared to match the composition of the basaltic glass produced in run B302 of Gaetani & Grove (1998), because it coexisted with a spinel lherzolite assemblage at P–T conditions similar to those of this study. Two other glasses (‘low-Ca glass’ and ‘low-Fe glass’) were prepared to produce similar bulk compositions after specific amounts of CaSO4 or FeSO4 were added (calculated S-free in Table 1). Thus, the bulk compositions of 10·76:1 mixtures (by weight) of ‘low-Ca glass’ + CaSO4 are similar to 9·54:1 mixtures of ‘low-Fe glass’ + FeSO4. Both glass–sulfate mixtures are roughly equal in bulk composition to a combination of the 70:30 basalt–peridotite (by weight) mixture used as starting material by Gaetani & Grove (1998) plus 5 wt % SO3 (Table 1).
|
All glasses were prepared from high-purity oxides and carbonates, which were ground and mixed thoroughly in an agate mortar under ethanol, partially decarbonated at 350°C for 4 h, fused in a Pt crucible at 1400°C for 30 min, and quenched by dipping the crucible in water. Pieces of the quenched glasses were inspected with a petrographic microscope at 200x for the presence of crystals. Partially crystalline glasses (containing small pieces of spinel) were ground again and fused at higher temperatures (in 25°C increments, up to 1450°C) until the quenched glasses were optically homogeneous and free of crystals, then ground and fused once more at the established melting temperatures. Random pieces of the resulting glass were mounted for electron microprobe analysis (EMPA) to test for chemical composition and homogeneity. The remaining glass was ground and stored in glass vials in a desiccator.
Sulfur
Sulfur was added as FeS, CaSO4, or FeSO4·H2O (ferrosulfate monohydrate). The ferrosulfate monohydrate was prepared by dehydrating ferrosulfate heptahydrate (FeSO4·7H2O) at 120°C. Amounts of ferrosulfate monohydrate equivalent to the desired amount of anhydrous ferrosulfate (FeSO4) were loaded into the capsules, which were heated at 350°C for 30–60 min immediately before welding to remove the remaining H2O.
Other phases
To promote crystallization of olivine, some runs were seeded with small amounts of finely crushed (<325 mesh), handpicked olivine crystals (KR-37, Fo91; composition shown in Table 1) from West Kettle River, British Columbia (Fujii & Scarfe, 1982; Canil et al., 1990). Some of the ‘sulfate-reduction’ experiments used small amounts of graphite powder as a reducing agent when run in Au–Pd capsules, and in a few experiments small amounts of Cr2O3 powder were mixed with the other starting materials to promote nucleation and growth of spinel crystals.
Bulk compositions
The first set of experiments (sulfide-saturated) used 5:1 mixtures (by weight) of B302 glass:FeS, with total masses ranging from 40·50 to 63·60 mg. The other two sets of experiments used various combinations of starting materials. Table 2 shows the type and amount of starting material added in these experiments (except precious metals) as well as the bulk composition of each run in terms of oxides.
|
Capsule materials
Graphite-lined platinum capsules were used for most of the sulfide-saturated experiments because the presence of graphite holds the oxidation state in this type of experiment at reducing conditions (Ulmer & Luth, 1991
; Holloway et al., 1992). The reducing effect of graphite prohibited its use in sulfate-saturated experiments, making the choice of capsule material the major limitation for these experiments. Unlined Pt capsules are unusable because Pt reacts with both Fe and S in the charge, resulting in Fe loss by formation of Fe–Pt alloy, and capsule corrosion by formation of PtS. Gold is effectively inert under these conditions but its relatively low melting point (1064°C at 1 atm) makes it unsuitable for the temperatures of this study (1300°C). Capsules made of Au–Pd alloys (Au75Pd25 and Au80Pd20, by weight) have sufficiently high melting points (>1350°C at 1 atm; Massalski, 1986). These alloys reacted slowly with S to form PdS, which weakened the alloy and limited the duration of experiments to approximately 6 h. We determined, however, that sulfate-saturated experiments reached equilibrium after only 2 h. Therefore, we used this capsule material in some of the sulfate-reduction experiments and all the sulfate-saturated experiments. Au–Pd capsules were used only at temperatures below 1325°C to provide a safety margin in temperature between the run conditions and the melting point of the alloy.
Experimental methods
For all the experiments, an end-loaded, solid-media, piston-cylinder apparatus (Boyd & England, 1960; Dunn, 1993) was used. The experiments used 19 mm diameter talc–Pyrex sample assemblies of the Kushiro (1976) type. The sample volume in these assemblies can accommodate one 5 mm outer diameter (o.d.) capsule up to 16 mm in length, and is large enough to accommodate two 3 mm o.d. capsules side by side, or two 5 mm o.d. capsules (and up to 8 mm length) on top of each other. Because of the constraints imposed by the Au–Pd capsules on the duration of the experiments, several paired experiments were run to test for equilibrium. Excess space in the sample volume (after loading the capsules) was filled in most experiments with crushed alumina powder. The use of crushed alumina as filler worked well in experiments using Pt capsules but Au–Pd capsules are more brittle, and broke easily when removed from the assembly at the end of the runs. Crushed Pyrex was tried as filler, but was rejected because Au–Pd capsules cracked upon quenching and molten Pyrex entered and mixed with the run products. The best results were obtained using sintered hematite as filler. The use of hematite as filler reduced significantly the cracking and fracturing of the Au–Pd capsules on quenching, probably because it did not attach as much as alumina to the capsules.
A Bourdon-tube Heise gauge with p.s.i. graduations (0–10 000 p.s.i., 0–68·9 MPa), with 10 p.s.i. (0·0689 MPa) graduations, was used to monitor line pressure. Run pressures were kept within ±0·14 MPa of the desired pressure in all experiments. No friction correction was applied to the nominal pressure because pressure calibration of the assemblies, based on the grossular + quartz = anorthite + 2 wollastonite equilibria (Windom & Boettcher, 1976), showed that the friction correction is a function of temperature and is negligible above 1275°C.
Type-C thermocouples constructed from batch-calibrated W5Re95 and W26Re74 wires (Omega Corp.) were used to measure run temperatures. At 1300°C, temperature uncertainties associated with variations in current are 1°C. Larger uncertainties are produced by thermal gradients along the capsule volume, which are 10°C for tapered graphite furnaces like those used in these experiments (Kushiro, 1976). The standard limits of error for type-C thermocouples at 1573 K (1300°C) are about 14°C (Dunn, 1993). Thus, the added uncertainties in temperature are about 25°C (2%). No correction was attempted for the effects of pressure or pressure gradients on the electromotive force of the thermocouples.
Analytical techniques
Elemental concentrations in all phases were determined by EMPA using wavelength-dispersive spectrometry (WDS) on a JEOL JXA-8900 electron microprobe. Typical operating conditions were 15 kV accelerating voltage and 15 nA probe current. Natural glasses and minerals (Jarosewich et al., 1980) were used as standards. Matrix effects were corrected using the ZAF method. Both barite and sphalerite were used as standards for S analyses. No significant difference was detected in the use of sphalerite or barite as standards for S concentrations in the sulfide-saturated glasses, but only barite was used as a standard for sulfate-bearing runs. Glasses were mostly analyzed with beams of 10 µm diameter, although 5 µm diameter beams were used in a few runs that contained only small melt pockets. Spinel occurred as very small crystals, with cross-sections rarely larger than 3 µm, making them difficult to analyze. They were analyzed with a focused beam (1 µm diameter) and with the inclusion of Si in the measurements as a monitor for the presence of silicates within the activation volume. Spinel crystals larger than 3 µm diameter contained <1 wt % SiO2; hence, only analyses with <1 wt % SiO2 were considered to be acceptable for spinel.
Control and estimation of oxidation state
Oxidation state in the experiments was constrained using combinations of starting and capsule materials. Control of oxidation state using external oxygen buffers was not possible because external buffers require the presence of a hydrous fluid phase in the charge, and water-saturated experiments in this system quench to a felt-like mixture of amphibole crystals, which means that melt compositions cannot be inferred. To impose reducing conditions, graphite-lined capsules (or small amounts of powdered graphite added to the starting materials) were used in combination with sulfide- and sulfate-bearing starting materials to produce sulfide-saturated assemblages. Conversely, oxidized starting materials (containing glasses fused in air, and sulfates) were enclosed in Au–Pd capsules to produce sulfate-saturated assemblages. Because the experiments were anhydrous and hematite was used as filler, little hydrogen diffused into the capsule from the assembly, and the oxidation state of the charges was dictated by the combination of starting materials used. Using this approach, we were able to produce runs equilibrated at fO2 above the nickel–bunsenite buffer (NNO, FMQ + 0·7) and below the manganosite–hausmannite buffer (MNH, FMQ + 4). However, we could not predetermine the oxidation state of the runs, nor obtain sulfate-saturated runs equilibrated below FMQ = +1·85.
The oxidation state of the experiments was estimated using the calibration of Ballhaus et al. (1990, 1991) for the olivine–orthopyroxene–spinel (ol–opx–spl) oxybarometer. This empirical calibration was used because Ballhaus et al. (1990) showed that it is more accurate over a wider range of oxidation states than other available calibrations, and also because it does not require the chemical composition of the orthopyroxene as an input parameter [Ballhaus et al. (1991) estimated that the correction required for orthopyroxene-undersaturated ultramafic rocks would rarely exceed 0·2 log fO2 units]. Olivine and spinel compositions were determined by EMPA, and the Fe3+ content of spinel was calculated assuming stoichiometry. Although estimations of oxidation states from EMPA of spinel have been regarded as inaccurate because of the errors associated with estimation of ferric iron (e.g. Wood & Virgo, 1989; Canil et al., 1990), Ballhaus et al. (1991) showed that the uncertainties in Fe3+ decrease with increasing fO2 because of the increase in the magnetite component in spinel. Parkinson & Arculus (1999) showed that errors in fO2 attributable to Fe3+ in spinel calculated from stoichiometry are significant (larger than 0·5 log units) for Fe3+/
Fe 
0·2 (corresponding to oxidation states <FMQ = –1), whereas for Fe3+/Fe 
0·2, the uncertainties associated with the stoichiometric estimation of Fe3+ become smaller, and are insignificant (<0·2 log units) above Fe3+/Fe 0·4 (approximately FMQ = +1).
To provide an independent check on the accuracy of EMPA of spinel for the estimations of oxidation state, we routinely analyzed two spinel samples (KR-35 and KR-37) for which Fe3+/Fe had previously been measured by Mössbauer spectroscopy (Canil et al., 1990). The calculated fO2 values from EMPA of both olivine and spinel using the oxybarometer calibration of Ballhaus et al. (1990, 1991) are FMQ = –0·35 ± 0·26 for KR-35 (from measurements compiled during four EMPA sessions) and FMQ = –1·07 ± 0·25 for KR-37 (from five sessions). The same calculations using the Mössbauer data for these two spinel samples (Canil et al., 1990) yield FMQ = –0·31 for KR-35 and FMQ = –0·86 for KR-37. Our results agree within error of these values, and thus justify the use of EMPA data for the calculations of both the Fe3+ content in spinel and the oxidation state of our experiments. The small discrepancies between our results and those documented by Canil et al. (1990) are more dependent on the choice of calibration of the oxybarometer (e.g. O'Neill & Wall, 1987; Mattioli & Wood, 1988; Ballhaus et al., 1990) than on the uncertainties associated with spinel analyses.
Estimation of equilibrium
Time invariance of the concentration of S in the quenched glasses was used as the criterion for equilibrium in sulfide-saturated experiments using graphite-lined Pt capsules. Experiments with duration ranging between 1 and 48 h had identical S concentrations, indicating that equilibrium was attained quickly (t 1 h). Time invariance could not be used as a test for equilibrium in oxidized experiments because of the limitations imposed by the reaction of S with Pd in Au–Pd capsules as mentioned above. Equilibrium in these experiments was monitored by comparing run products obtained from mixtures of ‘low-Ca glass’ + CaSO4 with run products obtained from mixtures of ‘low-Fe glass’ + FeSO4 as starting materials. Agreement in the composition of the quenched glasses and their S content is consistent with equilibrium being achieved in these runs.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The results of 22 successful experiments are included in this paper. Eight sulfide-saturated runs allow comparison with available data on the S content in sulfide-saturated basaltic melts (e.g. Haughton et al., 1974
; Poulson & Ohmoto, 1990). Nine sulfate-saturated runs provide the first experimental data on the S content in sulfate-saturated basaltic melts. Five sulfide-saturated (‘sulfate-reduction’) runs complement the other experiments and helped verify equilibration of the runs. Table 3 summarizes the run conditions and the products obtained in each experiment.
|
Phase compositions in run products
Glasses
The glasses generated in these experiments were picrobasaltic to basaltic in composition (Table 4). All of the glasses were very similar in composition, although glasses from oxidized experiments were slightly, but consistently, more mafic (MgO = 13·7 ± 1·4, n = 9) than those from sulfide-saturated runs (MgO = 11·8 ± 0·6, n = 13). The S concentration in sulfate-saturated glasses (S = 1·5 ± 0·2 wt %, n = 9) was
10 times higher than in sulfide-saturated glasses (S = 0·14 ± 0·02 wt %, n = 13).
|
Sulfides
Sulfide phases had rounded shapes in polished section, which is consistent with their presence as immiscible liquids at the run conditions. They were pyrrhotitic in composition with S = 36·9 ± 0·9 wt %, and Fe = 60·2 ± 1·2 wt % (n = 12).
Sulfates
Sulfates also quenched to rounded blebs (Fig. 1), which is regarded as evidence that sulfate was present as an immiscible liquid in the oxidized runs. Polishing of sulfate-saturated samples using water to remove abrading material between polishing steps resulted in intense dissolution of the sulfates. Sulfate blebs were best preserved if samples were polished using only oil and cleaned with methanol. However, even in these cases, irregular, uneven, and plucked surfaces resulted because of the low hardness of the sulfates relative to the surrounding glass. For this reason, quantitative analysis of the sulfates using WDS was not possible. The wavelength of the S K
radiation (Carroll & Rutherford, 1988) for the blebs agreed with the wavelength of the S K for the barite standard, confirming that the blebs were sulfates. Qualitative energy dispersive analyses (EDS) showed that the sulfates were invariably Ca- and S-rich with small amounts of Na and Mg, even in experiments in which S was added as FeSO4. This latter point is taken as further supporting evidence of equilibrium between the sulfates and the melt.
|
Olivine
Olivine typically formed relatively large (up to 100 µm length), euhedral to subhedral crystals. Individual crystals were compositionally homogeneous, with Fo content ranging from 0·90 to 0·95 (Table 5). To verify that olivine equilibrated with the melt, the partitioning of magnesium and ferrous iron between olivine and melt (KD, Roeder & Emslie, 1970
) was calculated. Ferrous iron content in the glasses was first calculated by subtracting the ferric iron content [estimated using the method of Kilinc et al. (1983)] from the total iron measured as FeO by EMPA. The average KD was 0·32 ± 0·4 for reduced experiments, and 0·31 ± 0·01 for oxidized runs (for which 0·26 Fe3+/Fe 0·49). These values are consistent with the expected KD (KD 0·33 at 1 GPa; Ulmer, 1989) and show that olivine was in equilibrium with the silicate melt.
|
Spinel
Spinel was present as very small (from <1 µm and rarely up to 5 µm) equant crystals with trapezoidal sections, which were included in olivine or isolated within quenched glasses. Their compositions are shown in Table 6.
|
Pyroxenes
Orthopyroxene and clinopyroxene were present in some experiments (Table 3). Both pyroxenes ranged in size from a few micrometers to 15 µm diameter. Orthopyroxene crystals were usually equant, whereas the habit of clinopyroxene ranged from equant to slightly prismatic. Clinopyroxene with cores of orthopyroxene was observed in some cases. Although the compositions of these phases are not considered further in this paper, Tables 7 and 8 list compositions of clinopyroxene and orthopyroxene analyzed from the experiments.
|
|
S concentration as a function of oxidation state
Figure 2 shows measured S concentrations plotted against estimated oxidation states in the experiments. Olivine was absent in the run products from the first set of sulfide-saturated experiments, which meant that the oxidation state could not be estimated using olivine–orthopyroxene–spinel equilibria. However, because these experiments used graphite-lined capsules, their fO2 was constrained to lie below the C–CO buffer (
FMQ < –1 at the P and T of the experiments). The rectangular box in Fig. 2 corresponds to the range in S concentration (width is ±1
standard deviation) and the approximate fO2 range of these runs.
|
The lowest fO2 in sulfate-saturated experiments was measured in run 34 (
FMQ = +1·85), which contained a small amount of graphite in the starting material but not enough to cause complete reduction of sulfate and saturation in sulfide (as in runs 54A and 54B). The highest fO2 in sulfide-saturated experiments was measured in run 27 (FMQ = –1·12). A drastic change in the S content occurs between the fO2 conditions of these two experiments (roughly in the range –1 < FMQ < +2). However, despite repeated attempts, we could not obtain results at intermediate fO2 (ideally from quenched glasses saturated and equilibrated simultaneously with both sulfide and sulfate liquids). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
To the best of our knowledge, the data shown here are the first experimental data for the S content of basaltic melts saturated with a sulfate phase. The most significant result from these experiments is the 10-fold increase in S content between sulfide- and sulfate-saturated basaltic melts at 1300°C and 1 GPa.
Experimental data on the S content of intermediate to felsic melts (ranging from trachyandesites to dacites) saturated in anhydrite were documented by Carroll & Rutherford (1985, 1987) and Luhr (1990). Although their experiments were conducted at lower pressures and temperatures [up to 290 MPa and 1025°C in Carroll & Rutherford (1987); up to 400 MPa and 1000°C in Luhr (1990)] the results showed that the S contents in sulfate-saturated melts were significantly higher than the S content in sulfide-saturated melts. Luhr (1990) extrapolated the increase in the S content at sulfate saturation with rising temperature in his experiments and concluded that ‘relatively oxidized (>NNO), vapor-undersaturated basaltic melts at Benioff zone conditions (T = 1200°C and P = 30 kbar) could contain 1·5 wt % SO3 and perhaps as much as 2·5 wt % SO3’ (Luhr, 1990, p. 1109). Our results exceed this prediction, showing that at 1300°C and 1 GPa (10 kbar) a basaltic magma can dissolve 1·5 ± 0·2 wt % S (3·75 ± 0·5 wt % SO3).
Sulfur solubility and the ‘sulfur solubility minimum’
A relevant aspect regarding the behavior of S in silicate melts is how the S content changes within the stability fields of sulfide and sulfates. This is relevant because results from experiments equilibrated with gas mixtures show a minimum in the S concentrations in the glasses at about FMQ = +1 (e.g. Fincham & Richardson, 1954; Nagashima & Katsura, 1973; Katsura & Nagashima, 1974). This minimum has been regarded in some cases as a ‘sulfur solubility minimum’ (e.g. Kress, 1997). Those experiments, however, were not saturated with either sulfide or sulfate (as condensed phases). Therefore, the S concentrations they define do not provide direct information about the S concentrations required to saturate a silicate melt in either sulfides or sulfates (crystalline or liquid). Nonetheless, those results (and the minimum in sulfur content) have applications to the behavior of sulfur in magmas containing a volatile phase after condensed S-bearing phases are eliminated.
In the strict sense, ‘solubility’ (e.g. for FeS) refers to the maximum amount of a phase (FeS) that a solvent phase (e.g. a silicate melt) can dissolve at a given P and T. The term ‘solubility’ is phase specific and therefore, for any given P, T and melt composition, the solubility of FeS is different from the solubility of CaSO4, or any other S-bearing phase (as shown in this work). Thus, in a very strict sense, ‘sulfur solubility’ would apply only to melts saturated with elemental S, which is seldom the case. The term ‘solubility’ is loosely used to refer to the sulfur content in melts saturated with a S-bearing phase; but because the nature of the saturating phase is critical, the saturating phase needs to be specified clearly and direct extrapolations of the results from one phase to the behavior of other S-bearing phases are not valid.
Experiments equilibrated with gas mixtures have the additional complication that the gases contain several species (e.g. S2, H2S, COS, SO2, SO3). According to Henry's law, the concentration of these species in the melt is proportional to their fugacities in the gas mixture. Thus, the total sulfur concentration in the melt is the contribution of the S concentrations imposed by each gas species. Carroll & Webster (1994) showed that for fO2 above FMQ = –1: (1) sulfur is dominantly present as SO2; (2) the fugacities of reduced species (S2, H2S, COS) decrease; (3) fSO3 increases. Consequently, the variations in the S content for gas-equilibrated experiments reflect mostly: (1) low SO2 solubility in the melts relative to the other S species; (2) the relative changes in the fugacities of these other S species.
When a silicate melt is simultaneously in equilibrium with a S-bearing condensed phase (e.g. FeS, CaSO4) and a S-bearing gas phase, the total S content in the melt results from their contribution. Thus, as long as sulfides (or sulfates) are present, the S content in the melt will not show the changes shown in gas-equilibrated experiments and no ‘sulfur solubility minimum’ would exist. This can be seen in Fig. 3, which compares the sulfur content at sulfide and sulfate saturation of trachyandesitic melts (Carroll & Rutherford, 1985, 1987) and basaltic melts (this work). The data for sulfide-saturated experiments from Carroll & Rutherford (1985, 1987) clearly show that S concentration in trachyandesitic melts at sulfide saturation remain constant at about 0·350 wt %, independent of changes in fO2. This was also demonstrated by O'Neill & Mavrogenes (2002), who showed that the sulfur concentration at sulfide saturation was independent of fO2.
|
Transition from sulfide to sulfate stability
The relevance to magmatic processes of the increase in the S content in basaltic melts from sulfide saturation to sulfate saturation depends on the fO2 at which sulfides oxidize to sulfates. For any given composition, this would be constrained by the lowest fO2 at which sulfates are stable and the highest fO2 at which sulfides are stable. The results of our experiments show that sulfates were stable at fO2 as low as
FMQ = +1·85, but the highest fO2 at which sulfides are stable is probably higher than that shown by our data (FMQ = –1·12), which is rather an indication of the fO2 imposed by the presence of graphite.
Extrapolation of the sulfide–sulfate boundary from other experimental work was not attempted because the effect of composition is significant. Stormer & Carmichael (1971) showed that silica undersaturation favors sulfide oxidation at a given fO2 because in reactions such as
 | (1) |
An analogous reaction, more appropriate for mantle assemblages, can be written as
 | (2) |
As in equation (1), a decrease in the activity of silica at a fixed fO2 promotes S dissolution in the silicate melt as SO3 (which eventually could result in the precipitation of sulfate-bearing phases) at the expense of the troilite component in the sulfides.
Carroll & Rutherford (1987) used a similar reaction,
 | (3) |
NNO = +1 (FMQ = +1·7) for reaction (3) by fixing the activities of all condensed phases to unity. Using this equilibrium they concluded that a decrease in silica activity would make sulfates stable at lower fO2 whereas a decrease in the activity of wollastonite would have the opposite effect.
In more complex systems, the activities of components in condensed phases will be different from unity and, therefore, fO2 for the transition from sulfide to sulfate will differ from those defined by reaction (1), (2) or (3) using pure components. Systematic experimental studies are needed to properly constrain the transition from sulfide to sulfate as a function of fO2, P, T, and melt composition. Estimates of the effect of composition can be made, for example, using experimental data from Carroll & Rutherford (1987) and Scaillet et al. (1998), in which trachyandesites and dacites (respectively) were used as starting materials. Experiments 142a and 141a of Carroll & Rutherford (1987) showed that, at 204 MPa and 927°C, anhydrite was stable at fO2 as low as FMQ = +1·39 and pyrrhotite was stable at fO2 as high as FMQ = +0·89 (Fig. 3). Experiment 42b of Scaillet et al. (1998) showed that anhydrite was present at fO2 as low as FMQ = +1·9 under similar P–T conditions (225 MPa and 899°C). Thus, it appears that the change in composition from dacite to trachyandesite lowers the fO2 needed to stabilize anhydrite by about 0·5 log fO2 units.
We speculate that for basaltic melts the change from sulfide to sulfate stability occurs at fO2 of about FMQ = +1·8 (based on our results) and that decrease in silica activity could push this to fO2 as low as FMQ = +1 as shown by Sisson (2003) for basanites from Hawaii (Fig. 3).
Implications for magmatic processes
The marked increase in the S content with increasing oxidation state has significant implications for the behavior of S during partial melting of the mantle and during magmatic evolution. Because highly siderophile elements (Rh, Ru, Pd, Re, Os, Ir, Pt, and Au) are dominantly hosted in sulfide phases (Mitchell & Keays, 1981), generation of silicate melts enriched in precious metals is favored by elimination of sulfides. Sulfide dissolution, either by large degrees of melting (e.g. Keays, 1995) or by two stages of melting (e.g. Wyborn & Sun, 1994), is the most commonly proposed mechanism for sulfide elimination.
The importance of oxidation as an alternative mechanism for sulfide elimination in primitive magmas has been discussed qualitatively (e.g. Richards et al., 1991; Sillitoe, 1997; Mungall, 2002). Sulfide elimination by oxidation of the mantle source is consistent with hypotheses that link Au-rich and high Au/Cu magmatic–hydrothermal ore deposits with oxidized magmas (e.g. Richards, 1995; Sillitoe, 1997). Parkinson & Arculus (1999) estimated that the fO2 of arc magmas at their source ranged from FMQ = +0·5 to FMQ = +1·7, and Blatter & Carmichael (1998) showed that fO2 in the subarc upper mantle can reach FMQ = +2·4. Thus, it can be inferred that in the most oxidized arc sources sulfide phases would not be stable. Under such conditions chalcophile and siderophile element behavior will be controlled by alloy–melt or oxide–melt equilibria, and some elements would probably be incompatible. Evidence from lherzolites and harzburgites from orogenic massifs indicates that ‘except for Pd and Au, the PGE reside in mantle sulfides as melting-resistant atomic clusters or micro alloys’ (Lorand et al., 1999, p. 957). Capobianco et al. (1994) showed that, in the absence of sulfides, Pd behaves as an incompatible element whereas Ru and Rh are compatible in oxide phases. Thus, magma generation in oxidized sources in which sulfides are not stable would fractionate Au and Pd from other siderophile elements. A low activity of silica in the melt (e.g. produced by low degrees of partial melting and relatively high alkali contents) would allow for sulfide elimination at lower oxidation states, which is consistent with the association of Au-rich deposits with oxidized alkalic magmas (e.g. Richards, 1995). An example of this was shown by Sisson (2003), who found a native gold grain in basanites from Hawaii. The fO2 estimated in these basanites is FMQ = +0·9, but this was sufficient to resorb magmatic sulfides and liberate gold hosted in them. In fact, Sisson (2003) found that the proportion of sulfate in the basanites increased with decreasing silica, rather than with increasing fO2.
Even if magmas are generated within the stability field of sulfides, many magmas become oxidized enough during ascent and evolution to preclude, or reverse, sulfide saturation. For example, Metrich et al. (1999) and de Hoog et al. (2001) documented S concentrations in basaltic glasses from subduction zones approaching 3000 ppm, and Wallace (2002) measured up to 6000 ppm S in olivine-hosted melt inclusions from cinder cones associated with Popocatepetl (Mexico). These S contents greatly exceed the S concentration predicted in sulfide-saturated basaltic melts but are within the range of S concentrations expected for oxidized systems as shown in this work. Ballhaus (1993) showed that the fO2 of island-arc magmas ranged from FMQ = +1 to FMQ = +3 and that back-arc basalts and island-arc basalts can reach FMQ = +2. Thus, at oxidation states above those at which sulfide phases are stable, S in the form of dissolved sulfate will behave as an incompatible element and its concentration in the melt will increase with crystallization and fractionation until: (1) sufficiently high concentrations are reached to precipitate anhydrite; (2) another sulfate-bearing phase (e.g. nosean, haüyne, apatite) saturates; or (3) a vapor phase (S-rich) exsolves.
The significant increase in S content of the melt between sulfide and sulfate saturation may also help explain high-S degassing in some volcanoes. Estimates (de Hoog et al., 2003) of the fO2 of basalts associated with the Pinatubo eruption (based on Cr-spinel-bearing olivine) yield NNO = +1·4 (roughly FMQ = +2·1) and are almost identical to those of the dacitic magma erupted (roughly FMQ = +2·3). These estimates indicate that the basaltic melt and the dacitic magma had similar high oxidation states. At such high oxidation states, no sulfide phases would be present and all available S should be dissolved in the melt (dominantly as sulfate species) or present as a sulfate phase (e.g. anhydrite). Because the S content at sulfate saturation decreases sharply with falling temperature and decreasing pressure (Carroll & Rutherford, 1987; Luhr, 1990) decompression and cooling of oxidized, S-rich basaltic melts as they ascend and mix with overlying hydrous dacitic magma to produce a hybrid andesite (Pallister et al., 1996) should result in saturation of anhydrite. However, sulfate partitions preferentially into a hydrous phase and water exsolution would strip the melt of most of its S (as SO2; Keppler, 1999). Because of the large amounts of S that can be dissolved in oxidized basaltic melts, relatively small amounts of basalt could provide large quantities of S to exsolved volatile phases.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The results of this work show that changes in fO2 have a significant impact on the behavior of S in basaltic systems, producing a 10-fold increase in the S content of basaltic magmas at high fO2. Thus, modeling of physico-chemical processes in which S-bearing phases are important (e.g. sulfide saturation and elimination, melt enrichment in precious metals, S-rich explosive volcanism) needs to consider both the range in fO2 in which sulfide oxidizes to sulfate and the increase in the S content related to this change.
|
ACKNOWLEDGEMENTS |
|---|
This research was supported by NSERC grants to R.W.L. and J.P.R. Additional financial support in the form of a Geological Society of America Student Research Grant, C. M. Scarfe Scholarships in Experimental Petrology from the University of Alberta, and a Canadian Association of Petroleum Producers Scholarship to P.J.J. are gratefully acknowledged. We thank R. J. Arculus and an anonymous reviewer for their comments, which helped improve this work. We also thank M. Wilson for the editorial management.
* Corresponding author. Present address: Institut für Mineralogie, J. W. Goethe Universität, D-60054, Frankfurt am Main, Germany. Telephone: ++49 69 798 23145. Fax: ++49 69 798 22101. E-mail: Pedro.Jugo{at}em.uni-frankfurt.de
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL AND ANALYTICAL...RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Ballhaus, C. (1993). Redox states of lithospheric and asthenospheric upper mantle. Contributions to Mineralogy and Petrology 114, 331–348.[CrossRef][Web of Science]
Ballhaus, C., Berry, R. F. & Green, D. H. (1990). Oxygen fugacity controls in the Earth's upper mantle. Nature 348, 437–440.[CrossRef]
Ballhaus, C., Berry, R. F. & Green, D. H. (1991). High-pressure experimental calibration of the olivine–orthopyroxene–spinel oxygen geobarometer: implications for the oxidation state of the upper mantle. Contributions to Mineralogy and Petrology 107, 27–40.[CrossRef][Web of Science]
Blatter, D. L. & Carmichael, I. S. E. (1998). Hornblende peridotite xenoliths from central Mexico reveal the highly oxidized nature of subarc upper mantle. Geology 26, 1035–1038.
Boyd, F. R. & England, J. L. (1960). Apparatus for phase-equilibrium measurements at pressures up to 50 kilobars and temperatures up to 1750°C. Journal of Geophysical Research 65, 741–748.
Brauns, R. (1914). Skapolithführende Auswürflinge aus dem Laacher Seegebiet. Neues Jahrbuch für Mineralogie, Geologie und Paläeontologie 39, 79–125.
Canil, D., Virgo, D. & Scarfe, C. M. (1990). Oxidation state of mantle xenoliths from British Columbia, Canada. Contributions to Mineralogy and Petrology 104, 453–462.[CrossRef][Web of Science]
Capobianco, C. J., Hervig, R. L. & Drake, M. J. (1994). Experiments on crystal/liquid partitioning of Ru, Rh and Pd for magnetite and hematite solid solutions crystallized from silicate melt. Chemical Geology 113, 23–43.[CrossRef][Web of Science]
Carroll, M. R. & Rutherford, M. J. (1985). Sulfide and sulfate saturation in hydrous silicate melts. Journal of Geophysical Research 90, C601–C612.
Carroll, M. R. & Rutherford, M. J. (1987). The stability of igneous anhydrite: experimental results and implications for S behavior in the 1982 El Chichón trachyandesite and other evolved magmas. Journal of Petrology 28, 781–801.
Carroll, M. R. & Rutherford, M. J. (1988). Sulfur speciation in hydrous experimental glasses of varying oxidation state: results from measured wavelength shifts of sulfur X-rays. American Mineralogist 73, 845–849.[Abstract]
Carroll, M. R. & Webster, J. D. (1994). Solubilities of sulfur, noble gases, nitrogen, chlorine, and fluorine in magmas. In: Carroll, M. R. & Holloway, J. R. (eds) Volatiles in Magmas. Mineralogical Society of America, Reviews in Mineralogy 30, 231–279.
De Hoog, J. C. M., Mason, P. R. D. & van Bergen, M. J. (2001). S and chalcophile elements in subduction zones: constraints from a laser ablation ICP-MS study of melt inclusions from Galunggung Volcano, Indonesia. Geochimica et Cosmochimica Acta 65, 3147–3164.[CrossRef][Web of Science]
De Hoog, J. C. M., Hattori, K. H. & Hoblitt, R. P. (2003). Oxidized, S-rich mafic magma at Mount Pinatubo, Philippines. (EGS–AGU 2003 Spring meeting abstracts.) Geophysical Research Abstracts 5, 01660.
De Silva, S. L. & Zielinski, XX. (1998). Global influence of the AD1600 eruption of Huaynaputina, Peru. Nature 393, 455–458.[CrossRef]
Dunn, T. (1993). The piston-cylinder apparatus. Mineralogical Association of Canada, Short Course Handbook 21, 39–94.
Fincham, C. J. B. & Richardson, F. D. (1954). The behaviour of sulphur in silicate and aluminate melts. Proceeding of the Royal Society of London, Series A 223, 40–61.[Abstract]
Fujii, T. & Scarfe, C. M. (1982). Petrology of ultramafic nodules from West Kettle River, near Kelowna, southern British Columbia. Contributions to Mineralogy and Petrology 80, 297–306.[CrossRef][Web of Science]
Gaetani, G. A. & Grove, T. L. (1998). The influence of water on melting of mantle peridotite. Contributions to Mineralogy and Petrology 131, 323–346.[CrossRef][Web of Science]
Gerlach, T. M., Westrich, H. R. & Symonds, R. B. (1996). Preeruption vapor in magma of the climactic Mount Pinatubo eruption; source of giant stratospheric S dioxide cloud. In: Newhall, C. G. & Punongbayan, R. S. (eds) Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines. Quezon City: Philippine Institute of Volcanology and Seismology; Seattle: University of Washington Press, pp. 415–433.
Gurenko, A. A. & Schmincke, H.-U. (1998). Petrology, geochemistry, S, Cl, and F abundances, and S oxidation state of sideromelane glass shards from Pleistocene ash layers north and south of Gran Canaria (ODP Leg 157). Contributions to Mineralogy and Petrology 131, 95–110.[CrossRef][Web of Science]
Gurenko, A. A. & Schmincke, H.-U. (2000). S concentrations and its speciation in Miocene basaltic magmas north and south of Gran Canaria (Canary Islands): constraints from glass inclusions in olivine and clinopyroxene. Geochimica et Cosmochimica Acta 64, 2321–2337.[CrossRef][Web of Science]
Hattori, K. H. (1993). High-S magma, a product of fluid discharge from underlying mafic magma: evidence from Mount Pinatubo, Philippines. Geology 21, 1083–1086.
Haughton, D. R., Roeder, P. L. & Skinner, B. J. (1974). Solubility of S in mafic magmas. Economic Geology 69, 451–466.
Holloway, J. R., Pan, V. & Gudmundsson, G. (1992). High-pressure fluid-absent melting experiments in the presence of graphite: oxygen fugacity, ferric/ferrous ratio and dissolved CO2. European Journal of Mineralogy 4, 105–114.
Holzheid, A. & Grove, T. L. (2002). Sulfur saturation limits in silicate melts and their implications for core formation scenarios for terrestrial planets. American Mineralogist 87, 227–237.
Jaques, A. L. & Green, D. H. (1980). Anhydrous melting of peridotite at 0–15 kbar pressure and the genesis of tholeiitic basalts. Contributions to Mineralogy and Petrology 73, 287–310.[CrossRef][Web of Science]
Jarosevich, E., Nelen, J. A. & Norberg, J. A. (1980). Reference samples for electron microprobe analysis. Geostandards Newsletter 4, 43–47.
Katsura, T. & Nagashima, S. (1974). Solubility of S in some magmas at 1 atmosphere. Geochimica et Cosmochimica Acta 38, 517–531.[CrossRef][Web of Science]
Keays, R. R. (1995). The role of komatiitic and picritic magmatism and S-saturation in the formation of ore deposits. Lithos 34, 1–18.[CrossRef][Web of Science]
Keppler, H. (1999). Experimental evidence for the source of excess S in explosive volcanic eruptions. Science 284, 1652–1654.
Kilinc, A. Carmichael, I. S. E., Rivers, M. L. & Sack, R. O. (1983). The ferric–ferrous ration of natural silicate liquids equilibrated in air. Contributions to Mineralogy and Petrology 83, 136–140.[CrossRef][Web of Science]
Kress, V. C. (1997). Magma mixing as a source for Pinatubo sulphur. Nature 389, 591–593.[CrossRef]
Kushiro, I. (1976). Changes in viscosity and structure of melt of NaAlSi2O6 composition at high pressures. Journal of Geophysical Research 81, 6347–6350.
Lindgren, W. & Ransome, F. L. (1906). Geology and gold deposits of the Cripple Creek district, Colorado. US Geological Survey Professional Paper 54.
Lorand, J.-P., Pattou, L. & Gros, M. (1999). Fractionation of platinum-group elements and gold in the upper mantle: a detailed study in Pyrenean orogenic lherzolites. Journal of Petrology 40, 957–981.[CrossRef][Web of Science]
Luhr, J. F. (1990). Experimental phase relations of water- and S-saturated arc magmas and the 1982 eruption of El Chichón volcano. Journal of Petrology 31, 1071–1114.
Luhr, J. F. & Varekamp, J. C. (1984). El Chichon volcano, Chiapas, Mexico (Introduction to the special volume). Journal of Volcanology and Geothermal Research 23, vii.[CrossRef]
Luhr, J. F., Carmichael, I. S. E. & Varekamp, J. C. (1984). The 1982 eruptions of El Chichón volcano, Chiapas, Mexico: mineralogy and petrology of the anhydrite-bearing pumice. Journal of Volcanology and Geothermal Research 23, 69–108.[CrossRef][Web of Science]
Massalski, T. B. (ed.) (1986). Binary Alloy Phase Diagrams. Metals Park, OH: American Society for Metals.
Matthews, S. J., Moncrieff, D. H. D. & Carroll, M. R. (1999). Empirical calibration of the sulphur valence oxygen barometer from natural and experimental glasses: method and applications. Mineralogical Magazine 63, 421–431.[Abstract]
Mattioli, G. S. & Wood, B. J. (1988). Magnetite activities across the MgAl2O4–Fe3O4 spinel join, with application to thermobarometric estimates of upper mantle oxygen fugacity. Contributions to Mineralogy and Petrology 98, 148–162.[CrossRef][Web of Science]
Mavrogenes, J. A. & O'Neill, H. St. C. (1999). The relative effects of pressure, temperature and oxygen fugacity on the solubility of sulfide in mafic magmas. Geochimica et Cosmochimica Acta 63, 1173–1180.[CrossRef][Web of Science]
McDonough, W. F. & Sun, S.-s. (1995). The composition of the Earth. Chemical Geology 120, 223–253.[CrossRef][Web of Science]
Metrich, N. & Clocchiatti, R. (1996). S abundance and its speciation in oxidized alkaline melts. Geochimica et Cosmochimica Acta 60, 4151–4160.[CrossRef][Web of Science]
Metrich, N., Schiano, P., Clocchiatti, R. & Maury, R. C. (1999). Transfer of S in subduction settings: an example from Batan Island (Luzon volcanic arc, Philippines). Earth and Planetary Science Letters 167, 1–14.[CrossRef][Web of Science]
Mitchell, R. H. & Keays, R. R. (1981). Abundance and distribution of gold, palladium and iridium in some spinel and garnet lherzolites: implications for the nature and origin of precious metal-rich intergranular components in the upper mantle. Geochimica et Cosmochimica Acta 45, 2425–2442.[CrossRef][Web of Science]
Mungall, J. E. (2002). Roasting the mantle: slab melting and the genesis of major Au and Au-rich Cu deposits. Geology 30, 915–918.
Nagashima, S. & Katsura, T. (1973). The solubility of S in Na2O–SiO2 melts under various oxygen partial pressures at 1100°C, 1250°C, and 1300°C. Bulletin of the Chemical Society of Japan 46, 3099–3103.[CrossRef]
Newhall, C. G., Daag, A. S., Delfin, F. G., Hoblitt, R. P., McGeehin, J., Pallister, J. S., Regalado, M. T., Rubin M., Tubianosa, B. S., Tamayo, R. A. & Umbal, J. V. (1996). Eruptive history of Mount Pinatubo. In: Newhall, C. G. & Punongbayan, R. S. (eds) Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines. Quezon City: Philippine Institute of Volcanology and Seismology; Seattle: University of Washington Press, pp. 165–195.
Nilsson, K. & Peach, C. L. (1993). S speciation, oxidation state, and S concentration in back-arc magmas. Geochimica et Cosmochimica Acta 57, 3807–3813.[CrossRef][Web of Science]
O'Neill, H. St. C. & Mavrogenes, J. A. (2002). The sulfide capacity and the S content at sulfide saturation of silicate melts at 1400°C and 1 bar. Journal of Petrology 43, 1049–1087.
O'Neill, H. St. C. & Wall, V. J. (1987). The olivine–orthopyroxene–spinel oxygen barometer, the nickel precipitation curve, and the oxygen fugacity of the Earth's upper mantle. Journal of Petrology 28, 1169–1191.
Pallister, J. S., Hoblitt, R. P., Meeker, G. P., Knight, R. J. & Siems, D. F. (1996) Magma mixing at Mount Pinatubo: petrographic and chemical evidence from the 1991 deposits. In: Newhall, C. G. & Punongbayan, R. S. (eds) Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines. Quezon City: Philippine Institute of Volcanology and Seismology; Seattle: University of Washington Press, pp. 687–731.
Palmer, A. S., van Ommen, T. D., Curran, M. A. J., Morgan, V., Souney, J. M. & Mayewski, P. A. (2001). High-precision dating of volcanic events (A.D. 1301–1995) using cores from Law Dome, Antarctica. Journal of Geophysical Research, D, Atmospheres 106, 28089–28095.[CrossRef]
Parkinson, I. J. & Arculus, R. J. (1999). The redox state of subduction zones: insights from arc-peridotites. Chemical Geology 160, 409–423.[CrossRef][Web of Science]
Poulson, S. R. & Ohmoto, H. (1990). An evaluation of the solubility of sulfide S in silicate melts from experimental data and natural samples. Chemical Geology 85, 57–75.[CrossRef][Web of Science]
Richards, J. P. (1995). Alkalic-type epithermal gold deposits—a review. In: Magmas, Fluids, and Ore Deposits, Mineralogical Association of Canada Short Course 23, 367–400.
Richards, J. P., McCulloch, M. T., Chappell, B. W. & Kerrich, R. (1991). Sources of metals in the Porgera gold deposit, Papua New Guinea: evidence from alteration, isotope, and noble metal geochemistry. Geochimica et Cosmochimica Acta 55, 565–580.[CrossRef][Web of Science]
Robinson, J. A. C., Wood, B. J. & Blundy, J. D. (1998). The beginning of melting of fertile and depleted peridotite at 1·5 GPa. Earth and Planetary Science Letters 155, 97–111.[CrossRef][Web of Science]
Roeder, P. L. & Emslie, R. F. (1970). Olivine–liquid equilibrium. Contributions to Mineralogy and Petrology 29, 275–289.[CrossRef][Web of Science]
Scaillet, B., Clemente, B., Evans, B. W. & Pichavant, M. (1998). Redox control of sulfur degassing in silicic magmas. Journal of Geophysical Research 103, 23937–23949.[CrossRef][Web of Science]
Self, S., Zhao, J., Holasek, R. E., Torres, R. C. & King, A. J. (1996). The atmospheric impact of the 1991 Mount Pinatubo eruption. In: Newhall, C. G. & Punongbayan, R. S. (eds) Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines. Quezon City: Philippine Institute of Volcanology and Seismology; Seattle: University of Washington Press, pp. 1098–1115.
Sillitoe, R. H. (1997). Characteristics and controls of the largest porphyry copper–gold and epithermal gold deposits in the circum-Pacific region. Australian Journal of Earth Sciences 44, 373–388.[Web of Science]
Sisson, T. W. (2003). Native gold in a Hawaiian alkalic magma. Economic Geology 95, 643–648.
Stormer, J. C. & Carmichael, I. S. E. (1971). The free energy of sodalite and the behavior of chloride, fluoride and sulfate in silicate magmas. American Mineralogist 56, 292–306.[Web of Science]
Stothers, R. B. (1984). The great Tambora eruption in 1815 and its aftermath. Science 224, 1191–1198.
Ulmer, P. (1989). The dependence of the Fe2+–Mg cation-partitioning between olivine and basaltic liquid on pressure, temperature and composition: an experimental study to 30 kbars. Contributions to Mineralogy and Petrology 101, 261–273.[CrossRef][Web of Science]
Ulmer, P. & Luth, R. W. (1991). The graphite–COH fluid equilibrium in P, T, fO2 space: an experimental determination to 30 kbar and 1600°C. Contributions to Mineralogy and Petrology 106, 265–272.[CrossRef][Web of Science]
Wallace, P. J. (2002). Volatile constituents and vapor saturation: integrating petrological and remote sensing perspectives to understand magmatic volatile budgets. EOS Transactions, American Geophysical Union 83(47), abstract V52D-13.
Windom, K. E. & Boettcher, A. L. (1976). The effect of reduced activity of anorthite on the reaction grossular + quartz = anorthite + wollastonite; a model for plagioclase in the Earth's lower crust and upper mantle. American Mineralogist 61, 889–896.[Abstract]
Wood, B. J. & Virgo, D. (1989). Upper mantle oxidation state: ferric iron contents of lherzolite spinels by 57Fe Mössbauer spectroscopy and resultant oxygen fugacities. Geochimica et Cosmochimica Acta 53, 1277–1291.[CrossRef][Web of Science]
Wyborn, D. & Sun, S.-S. (1994). Sulphur-undersaturated magmatism—a key factor for generating magma-related copper–gold deposits: AGSO Research Newsletter (November), 7–8.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. J. Jugo Sulfur content at sulfide saturation in oxidized magmas Geology, May 1, 2009; 37(5): 415 - 418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Li and E. M. Ripley Sulfur Contents at Sulfide-Liquid or Anhydrite Saturation in Silicate Melts: Empirical Equations and Example Applications Economic Geology, May 1, 2009; 104(3): 405 - 412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Li, E. M. Ripley, and A. J. Naldrett A NEW GENETIC MODEL FOR THE GIANT Ni-Cu-PGE SULFIDE DEPOSITS ASSOCIATED WITH THE SIBERIAN FLOOD BASALTS Economic Geology, March 1, 2009; 104(2): 291 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Scheel, J. S. Scoates, and G. T. Nixon CHROMIAN SPINEL IN THE TURNAGAIN ALASKAN-TYPE ULTRAMAFIC INTRUSION, NORTHERN BRITISH COLUMBIA, CANADA Can Mineral, February 1, 2009; 47(1): 63 - 80. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Rowe, A. J. R. Kent, and R. L. Nielsen Subduction Influence on Oxygen Fugacity and Trace and Volatile Elements in Basalts Across the Cascade Volcanic Arc J. Petrology, January 7, 2009; (2009) egn072v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Chambefort, J. H. Dilles, and A. J.R. Kent Anhydrite-bearing andesite and dacite as a source for sulfur in magmatic-hydrothermal mineral deposits Geology, September 1, 2008; 36(9): 719 - 722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Mungall and A. J. Naldrett Ore Deposits of the Platinum-Group Elements Elements, August 1, 2008; 4(4): 253 - 258. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Metrich and P. J. Wallace Volatile Abundances in Basaltic Magmas and Their Degassing Paths Tracked by Melt Inclusions Reviews in Mineralogy and Geochemistry, January 1, 2008; 69(1): 363 - 402. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Wilke, P. J. Jugo, K. Klimm, J. Susini, R. Botcharnikov, S. C. Kohn, and M. Janousch The origin of S4+ detected in silicate glasses by XANES American Mineralogist, January 1, 2008; 93(1): 235 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Webster and C. W. Mandeville Fluid Immiscibility in Volcanic Environments Reviews in Mineralogy and Geochemistry, July 1, 2007; 65(1): 313 - 362. [Full Text] [PDF]
|
|
|
|
|
|
J. P. Richards and R. Kerrich Special Paper: Adakite-Like Rocks: Their Diverse Origins and Questionable Role in Metallogenesis Economic Geology, June 1, 2007; 102(4): 537 - 576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Richards, D. Wilkinson, and T. Ullrich Geology of the Sari Gunay Epithermal Gold Deposit, Northwest Iran Economic Geology, December 1, 2006; 101(8): 1455 - 1496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Mungall, J. J. Hanley, N. T. Arndt, and A. Debecdelievre Evidence from meimechites and other low-degree mantle melts for redox controls on mantle-crust fractionation of platinum-group elements PNAS, August 22, 2006; 103(34): 12695 - 12700. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. SCAILLET and R. MACDONALD Experimental and Thermodynamic Constraints on the Sulphur Yield of Peralkaline and Metaluminous Silicic Flood Eruptions J. Petrology, July 1, 2006; 47(7): 1413 - 1437. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||