High sulfur solubility in subducted sediment melt under both reduced and oxidized conditions: With implications for S recycling in subduction zone settings
Introduction
Sulfur is one of the major volatiles released during volcanic processes. It controls the partitioning, transport and enrichment of ore metals during magmatic evolution and ore deposit formation. The subduction related S cycle has gained research attention in recent years due to its potentially important role in the evolution of mantle redox state (e.g., Kelley and Cottrell, 2009, Evans, 2012) and the formation of porphyry copper deposits (e.g., Richards, 2015, Canil and Fellows, 2017). Although estimates of S fluxes indicate a significant percentage of S is returned to the deep mantle (Wallace, 2005), the higher S content in arc basalt compared to MORB (Wallace, 2005) and high δ34S values in arc magma (e.g., De Hoog et al., 2001) suggest recycling of seawater-derived S from subducted slab to the mantle wedge. The mechanism through which such S transfer takes place remains poorly constrained. If S transfer was achieved via fluids or melts produced by slab devolatilization or fluxed-melting, the efficiency of S recycling would be dependent upon the S solubility,1 which, in turn, is a function of pressure (P), temperature (T), fluid/melt composition and oxygen fugacity (fO2).
Experimental studies of sulfur solubility in silicate melts, namely the sulfur content at sulfide saturation (SCSS) and the sulfur content at anhydrite saturation (SCAS), span several decades (Baker and Moretti, 2011), with a general focus on mantle melting and volcanic processes. More recently, a small number of studies (Prouteau and Scaillet, 2013, Jégo and Dasgupta, 2013, Jégo and Dasgupta, 2014, Canil and Fellows, 2017, Chowdhury and Dasgupta, 2019) have been devoted to the behavior of S during subduction processes.
Prouteau and Scaillet (2013) reported S solubilities in hydrous felsic melts produced by melting of hydrous MORB and pelitic compositions at 2–3 GPa, 700–950 °C and fO2 from NNO-2 to NNO + 1, with values falling in a large range of 100 ppm to 1 wt%, showing positive correlations with P, T, fO2 and bulk S content (1–4 wt%). Jégo and Dasgupta, 2014, Jégo and Dasgupta, 2013 studied the melting of hydrated basaltic crust with 1 wt% S in the PT range of 2–3 GPa, 700–1050 °C, and fO2 conditions from FMQ-3 to FMQ + 3.5. They reported consistently low SCSS values at reduced conditions (fO2 < FMQ) averaging ∼100 ppm, with higher values up to 518 ppm at intermediate conditions (FMQ < fO2 < FMQ + 2), and SCAS values of 700–3000 ppm at oxidized conditions (fO2 > FMQ + 2). While Prouteau and Scaillet (2013) reported S solubilities for hydrous felsic melts high enough to efficiently transfer S from the slab to the mantle wedge, particularly under relatively oxidized conditions, Jégo and Dasgupta, 2014, Jégo and Dasgupta, 2013 suggested SCSS values for rhyolitic melts are too low for them to be suitable carriers of S, and that aqueous fluids present a more efficient medium for S transfer. While the solubility data from the above two studies seem comparable at first order, their contrasting conclusions reflect the difficulty in constraining S recycling mechanisms based on the sparse experimental data currently available for limited P-T-composition-fO2 space. Generally speaking, S recycling in subduction zone settings has the potential to occur at different depths via differing mechanisms (Tomkins and Evans, 2015), accomplished through the mediums of aqueous fluid, silicate melt and supercritical fluids at different stages.
Further discrepancies arise when comparing experimental data from these studies with predictions using the published S solubility models of Clemente et al., 2004, Liu et al., 2007. Based on S solubility data in hydrous rhyolitic melts at low pressure (2 kbar), Clemente et al. (2004) developed a thermodynamic model relating S content in melt to fH2S and fSO2, in addition to an empirical model for S solubility in hydrous rhyolitic melts as a function of T, fO2 and fS2 when fO2 < NNO + 1.5. Liu et al. (2007) developed an empirical SCSS model based on experimental data from the literature covering a wide range of PT and melt compositions, however, this model was unable to reproduce the experimental data of Clemente et al. (2004). Empirical SCSS models suffer from sampling bias, as the majority of experimental data were obtained for anhydrous mafic melts. Wykes et al. (2015) demonstrated that the Liu et al. (2007) model underestimates the SCSS values for melt compositions with FeO contents <5 wt%. The dependence of SCSS on FeO content in melt can be depicted with a U-shaped curve, with SCSS positively correlated with FeO content (>5 wt%) in the right limb, while negatively correlated with FeO (<5 wt%) in the left limb (Haughton et al., 1974, O’Neill and Mavrogenes, 2002, Wykes et al., 2015). Prouteau and Scaillet (2013) reported that while their solubility data can be reproduced by the model of Clemente et al. (2004), it is significantly underestimated by the Liu et al. (2007) model. Conversely, Jégo and Dasgupta, 2014, Jégo and Dasgupta, 2013 report SCSS data that fit with model predictions of Liu et al. (2007).
In order to resolve such a discrepancy and explore the suitability of different S solubility models for high pressure melts produced by the fluxed melting of subducted slab, we performed a series of subducted sediment melting experiments at 3 GPa, 1050 °C with varying fO2 to observe the variations of SCSS and SCAS. The choice of PT conditions was intended for comparison with the 3 GPa, 1050 °C experimental series from Jégo and Dasgupta (2014). While 1050 °C is at the upper limit for sediment melting at sub-arc depth (e.g., Syracuse et al., 2010, Cooper et al., 2012), this high T enables the generation of large mineral-free melt pools in the experimental charge and facilitates S speciation analysis using Raman spectroscopy. Repetitive experiments were performed both at lower T and with higher bulk S content to assess related T and concentration effects.
Section snippets
Experimental starting materials
A synthetic experimental pelite starting material (EPSM, Table 1), with a major element composition similar to both the “Global Subducting Sediment” (GLOSS, Plank and Langmuir, 1998) and upper continental crust (Rudnick and Gao, 2003), with the addition of S, was chosen for the sediment melting experiments conducted in this study. Starting compositions were synthesized following the procedure previously outlined in Li and Hermann, 2015, Li and Hermann, 2017 for S-free EPSM compositions. Major
Phase relations and compositions
For 1050 °C reduced experiments (fO2 < FMQ), a simple phase assemblage of quenched melt, pyrrhotite and small fractions of the liquidus phases garnet and coesite were observed (Table 2; Fig. 2a); while the degree of melting was much lower for oxidized experiments (fO2 ≥ FMQ + 1.6), for which the mineral assemblage is garnet + coesite + anhydrite + kyanite + rutile (Table 2; Fig. 2b, c). For exp. LMD704 (FMQ + 1.6), we observed the coexistence of pyrrhotite and anhydrite (Fig. 2c). A small
Experimental equilibrium and the effect of carbon
Within such an experimental study, assuring the attainment of both compositional and fO2 equilibrium remains a primary consideration. Without conducting reversed or bracketed experiments, an equilibrium state is demonstrated by the homogeneity of phase relations and compositions. All experiments, other than LMD719, displayed phase homogeneity. With a 24-hour run time, the IrIrO-buffered exp. LMD719 preserved a two-layer structure (Fig. 2f), with the outer layer in fO2 equilibrium with the
Implications for S recycling in subduction zones
The average S content in arc basalt is estimated to be 1500 ± 500 ppm (Jégo and Dasgupta, 2013), higher than the average S content of ∼1000 ppm in MORB. The average H2O content in arc basalt is estimated to be ∼4 wt% (Plank et al., 2013). As a rough estimate, this 4 wt% H2O content and apparent excess of 500 ppm S in arc basalt, may be sourced from a subduction component with a H2O/S ratio of 80. The required composition of such a component may theoretically span a range from aqueous fluid
Conclusions
We have conducted sediment melting experiments at 3 GPa, 1050 °C and variable fO2 to investigate the variation of SCSS and SCAS with fO2 in hydrous rhyolitic melt at high pressure. We observe an increase of SCSS with decreasing fO2 from ∼200 ppm at FMQ-0.5 to ∼1800 ppm at FMQ-7.5, which correlates with the decrease in melt FeO from 3.5 to 0.3 wt%. The S content in melt increases sharply during the S2− to S6+ transition beginning at ∼FMQ, reaching a maximum value of ∼2600 ppm at FMQ + 1.6, when
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work benefited from discussions with Hugh St.C. O’Neill from the Australian National University and Joerg Hermann from the University of Bern. The authors would also like to thank Yunlu Ma, Yichuan Wang and Wei Yan from the High Pressure and High Temperature laboratory, Peking University for their assistance with the experimental program; Xiaoli Li from Peking University and Quan Wei from the Chinese Academy of Geological Sciences for their help with electron microprobe analyses. Comments
References (80)
- et al.
Stabilities of laurite RuS2 and monosulfide liquid solution at magmatic temperature
Chem. Geol.
(2004) - et al.
Solubility, speciation and dissolution mechanisms for CO2 in melts on the NaAlO2–SiO2 join
Geochim. Cosmochim. Acta
(1999) Raman microspectrometry of fluid inclusions
Lithos
(2001)- et al.
Sulphide–sulphate stability and melting in subducted sediment and its role in arc mantle redox and chalcophile cycling in space and time
Earth Planet. Sci. Lett.
(2017) - et al.
Effect of sulfate on the basaltic liquidus and Sulfur Concentration at Anhydrite Saturation (SCAS) of hydrous basalts – Implications for sulfur cycle in subduction zones
Chem. Geol.
(2019) - et al.
CO2 solubility and speciation in rhyolitic sediment partial melts at 1.5-3.0GPa - Implications for carbon flux in subduction zones
Geochim. Cosmochim. Acta
(2014) The redox budget of subduction zones
Earth-Science Rev.
(2012)- et al.
The effect of water on the sulfur concentration at sulfide saturation (SCSS) in natural melts
Geochim. Cosmochim. Acta
(2015) - et al.
Experimental determination of activities of FeO and Fe2O3 components in hydrous silicic melts under oxidizing conditions
Geochim. Cosmochim. Acta
(2003) - et al.
Accessory phase control on the trace element signature of sediment melts in subduction zones
Chem. Geol.
(2009)
Sulfur isotope systematics of basaltic lavas from Indonesia: Implications for the sulfur cycle in subduction zones
Earth Planet. Sci. Lett.
The system C-O in equilibrium with graphite at high pressure and temperature: An experimental study
Geochim. Cosmochim. Acta
Fluid-present melting of sulfide-bearing ocean-crust: Experimental constraints on the transport of sulfur from subducting slab to mantle wedge
Geochim. Cosmochim. Acta
Sulfur K-edge XANES analysis of natural and synthetic basaltic glasses: Implications for S speciation and S content as function of oxygen fugacity
Geochim. Cosmochim. Acta
The distribution of sulfur between haplogranitic melts and aqueous fluids
Geochim. Cosmochim. Acta
The dissolution mechanism of sulphur in hydrous silicate melts. II: Solubility and speciation of sulphur in hydrous silicate melts as a function of fO 2
Chem. Geol.
The dissolution mechanism of sulphur in hydrous silicate melts. I: Assessment of analytical techniques in determining the sulphur speciation in iron-free to iron-poor glasses
Chem. Geol.
Apatite as an indicator of fluid salinity: An experimental study of chlorine and fluorine partitioning in subducted sediments
Geochim. Cosmochim. Acta
Redox processes in subducting oceanic crust recorded by sulfide-bearing high-pressure rocks and veins (SW Tianshan, China)
Contrib. to Mineral. Petrol.
Sulfur concentration at sulfide saturation (SCSS) in magmatic silicate melts
Geochim. Cosmochim. Acta
Omphacite-bearing calcite marble and associated coesite-bearing pelitic schist from the meta-ophiolitic belt of Chinese western Tianshan
J. Asian Earth Sci.
Anhydrite solubility in differentiated arc magmas
Geochim. Cosmochim. Acta
Raman studies of sulfur in borosilicate waste glasses: Sulfate environments
J. Non. Cryst. Solids
Sulfur solubility in reduced mafic silicate melts: Implications for the speciation and distribution of sulfur on Mercury
Earth Planet. Sci. Lett.
Activity coefficients at low dilution of CrO, NiO and CoO in melts in the system CaO-MgO-Al2O3-SiO2 at 1400°C: Using the thermodynamic behaviour of transition metal oxides in silicate melts to probe their structure
Chem. Geol.
Why do mafic arc magmas contain ∼4wt% water on average?
Earth Planet Sci. Lett.
The chemical composition of subducting sediment and its consequences for the crust and mantle
Chem. Geol.
The oxidation state, and sulfur and Cu contents of arc magmas: implications for metallogeny
Lithos
Composition of the Continental Crust
Element recycling from subducting slabs to arc crust: A review
Lithos
The global range of subduction zone thermal models
Phys. Earth Planet. Inter.
Formation of abiotic hydrocarbon from reduction of carbonate in subduction zones: Constraints from petrological observation and experimental simulation
Geochim. Cosmochim. Acta
Melting of carbonated pelites at 2.5–5.0 GPa, silicate–carbonatite liquid immiscibility, and potassium–carbon metasomatism of the mantle
Earth Planet. Sci. Lett.
Separate zones of sulfate and sulfide release from subducted mafic oceanic crust
Earth Planet. Sci. Lett.
A thermodynamic study of pyrite and pyrrhotite
Geochim. Cosmochim. Acta
The effect of carbonates on near-solidus melting of pelite at 3GPa: Relative efficiency of H2O and CO2 subduction
Earth Planet. Sci. Lett.
Volatiles in subduction zone magmas: concentrations and fluxes based on melt inclusion and volcanic gas data
J. Volcanol. Geotherm. Res.
An accurate model to predict sulfur concentration at anhydrite saturation in silicate melts
Geochim. Cosmochim. Acta
Modeling the solubility of sulfur in magmas: A 50-year old geochemical challenge
Rev. Mineral. Geochemistry
Thermochemical Data of Pure Substances
Cited by (9)
Anhydrite solubility enhanced by CaO in silicate melts: Implications for sulfur cycling in subduction zones
2023, Geochimica et Cosmochimica ActaFavorable P–T–ƒO<inf>2</inf> conditions for abiotic CH<inf>4</inf> production in subducted oceanic crusts: A comparison between CH<inf>4</inf>-bearing ultrahigh- and CO<inf>2</inf>-bearing high-pressure eclogite
2022, Geochimica et Cosmochimica ActaCitation Excerpt :Previous studies have shown that prograde fluid-induced dissolution of carbonate and carbonaceous materials may represent an essential process by which C-bearing fluids are released from subducting slabs (Frezzotti et al., 2011; Ague and Nicolescu, 2014; Facq et al., 2014; Kelemen and Manning, 2015; Menzel et al., 2020; Tumiati et al. 2020; Vitale et al., 2020a). However, whether slab fluids are dominated by reduced (e.g., CH4, H2S, HS−) or oxidized (e.g., CO2, HSO4−, SO42−) species remains controversial, and this dynamic has significant implications for constraining the redox evolution of subducting slabs, mantle wedges, and related volcanic arcs (e.g., Debret and Sverjensky, 2017; Gerrits et al., 2019; Walters et al., 2019; 2020; Vitale et al., 2017; Piccoli et al., 2019; Li et al., 2020; Li et al., 2021, 2022). Determining the mobility and speciation of carbon-bearing phases in subduction zone fluids remains an active and challenging research frontier (e.g., Debret et al., 2021).
Sulfur dissolution capacity of highly hydrated and fluid-saturated dacitic magmas at the lower crust and implications for porphyry deposit formation
2022, Geochimica et Cosmochimica ActaCitation Excerpt :High H2O and fO2 can shift reactions (2) and (3) to the right, thus elevating the S content in the melts and fluids. Developing a unique model to depict the behavior of S in the magma-hydrothermal system is challenging because sulfide/anhydrite solubility in melt and fluid is a complex function of multiple variables, including temperature, pressure, melt composition, sulfide composition, S species, and valence states (e.g., Chowdhury and Dasgupta, 2019; Fortin et al., 2015; Jugo et al., 2010; Li et al., 2021; Masotta and Keppler, 2015; Matjuschkin et al., 2016; Nash et al., 2019; O’Neill and Mavrogenes, 2002; Smythe et al., 2017; Xu and Li, 2021; Zajacz, 2015; Zajacz and Tsay, 2019). Our experiments at a fixed pressure (1.0 GPa) and temperature (950 °C), but variable H2O and fO2 obtained three important results (Figs. 7 and 8): (1) At a given fO2, the S content in the dacitic melt increases with increasing melt hydration.
A thermodynamic model for sulfur content at sulfide saturation (SCSS) in hydrous silicate melts: With implications for arc magma genesis and sulfur recycling
2022, Geochimica et Cosmochimica ActaCitation Excerpt :As a safe measure, the database used for model calibration is comprised of experiments with fO2 conditions of FMQ and below. Additional datasets are included from Scaillet and MacDonald (2006), Matjuschkin et al. (2016), D’Souza and Canil (2018) and Li et al. (2021). In the database of Moretti and Baker (2008), data for hydrous silicate melts are from Bradbury (1983), Carroll and Rutherford (1985), Carroll and Rutherford (1987), Luhr (1990) and Clemente et al. (2004).
The use of boron nitride to impose reduced redox conditions in experimental petrology
2023, American MineralogistSupercritical fluid in deep subduction zones as revealed by multiphase fluid inclusions in an ultrahigh-pressure metamorphic vein
2023, Proceedings of the National Academy of Sciences of the United States of America