April 2017 Issue Table of Contents
Sulfides are a major potential repository for magmatic metals and sulfur. In relatively reduced magmas, there may be a dynamic interplay between sulfide liquids and magma degassing as magmas ascend/erupt. Sulfide-bubble aggregates may segregate to shallow levels. Exsolved fluids may oxidize sulfides to produce SO2 gas and metals, which can vent to the atmosphere with chalcophile metal ratios reflecting those in their parent sulfide liquids. Sulfide breakdown and/or sequestration timing and balance define the role of sulfides in both ore formation and the environmental impacts of volcanic eruptions, including during the evolution of large igneous provinces, which are key periods of heightened volcanism during Earth history.
Keywords: sulfide, outgassing, metals, partitioning, vapour, eruptions
Sulfides are a common feature of near-surface magmas (Fig. 1), and they play an important role in volcanic systems in the supply of sulfur and chalcophile metals to the atmosphere and to sites of ore formation. Sulfur is one of the most abundant and important volatile species produced by volcanic activity. Volcanic eruptions may produce large clouds of sulfur dioxide which, when injected into the stratosphere, convert to sulfate aerosols and may impact climate by absorbing incoming solar radiation and scattering it back into space. Tropospheric plumes of sulfur gases and aerosol may be large enough to cause environmental damage and health hazards. Sulfur-rich fluids at submarine mid-ocean ridges, formed by the outgassing of ascending basaltic melts, support sulfide-oxidising microbial life and modulate ocean chemistry and oxidation state.
Sulfur is a ubiquitous component of magmas and displays complex behaviour due to its ability to exist in many valence states and species (S2−, S6+, S0 and as S2, SO2, SO3 and H2S in the gas phase). Its behaviour in magmas is largely dependent on magma oxidation state (the availability of oxygen) (Carroll and Rutherford 1985). Under reduced conditions, sulfur dissolves as sulfide (S2−), and under oxidized conditions, as sulfate (S6+ in SO42−); under intermediate conditions, both speciation states are present. Dissolved sulfur species will progressively concentrate in melts during crystallization (of non-sulfur-bearing phases), until eventually the melt concentration of sulfur may reach the level required for the precipitation of (or ‘saturation in’) a non-volatile, sulfur-bearing phase. The form of this phase is dependent on both the fugacities (or put simply, the abundance or availability) of oxygen and sulfur in the system. Under oxidized conditions, the solid sulfur-bearing phase is anhydrite (CaSO4). In the relatively reduced magmas that are typical of mid-ocean ridges, for example, this sulfur-bearing phase might be monosulfide solid solution (mss) or an Fe–O–S immiscible liquid that can quench to a sulfide solid solution (Parat et al. 2011). Photomicrographs to show quenched sulfide liquids in a range of volcanic rocks, in the form of inclusions in crystals and in matrix glass, are shown in Figure 1.
Sulfide liquids display a range of compositions and may contain appreciable concentrations of metals, such as copper (Cu) and nickel (Ni). Sulfide liquids are dense relative to silicate melts and may settle gravitationally and be reworked multiple times, leading to the segregation of massive volcanigenic sulfide deposits (Ripley and Li 2013): these deposits may have substantial economic value. In volcanic systems, however, which involve magmas stored at low pressures, hydrothermal fluids may also coexist with sulfide-saturated magmas. Interaction between oxidising, water-rich fluids and sulfide liquids may promote an interplay between volcanic outgassing and sulfide saturation such that sulfides might break down, supplying sulfur directly to the atmosphere (Nadeau et al. 2010). Volcanoes are significant sources of metals to the atmosphere via gas and aerosol phases (Mather et al. 2012); where sulfide saturation occurs, some of these metals may derive directly from sulfide breakdown (Larocque et al. 2000).
Sulfur is also volatile in silicate melts, it partitions strongly into a vapour phase at low pressures in the crust, and sulfide saturation may have a key modulating effect on melt–vapour partitioning. Sulfur partitioning behaviour is well understood for a wide range of oxidation states and melt compositions. In general, experiments indicate that sulfur partitions strongly into the gas phase. This is particularly the case for more reducing conditions below the sulfate–sulfide transition because of the lower solubility of sulfur when it exists as the S2− ion than when it occurs dissolved as sulfate (S6+) under more oxidizing conditions. Saturation of the melt in sulfide or, under more oxidizing conditions, in anhydrite limits the sulfur concentration in the co-existing gas phase to just a few per cent by volume (Zajacz et al. 2012).
The interplay between and timing of these partitioning processes (between silicate melt, vapour and sulfide) has consequences for outgassing or sequestration in reduced volcanic systems. There are a range of possible processes involving sulfide saturation and vapour saturation of melts, and their consequences are summarised in Figure 2. The mass budget of sulfur available for degassing into the atmosphere in volcanic systems is often estimated using the sulfur concentration in melt inclusions (tiny increments of melt trapped in growing crystals), the so-called petrological method (Sigurdsson et al. 1985). However, only rarely is the sulfur in the sulfide liquid phase taken into account, beyond, that is, the recognition of sulfide saturation. The timing of sulfide saturation and vapour saturation in volcanic systems is likely critical for determining the extent to which metal-rich sulfides are sequestered and for determining the magnitude of sulfur-rich gas clouds outgassed with the magma during eruptions. The high degree of wetting of hydrous vapour on sulfide liquid droplets may render the sulfide liquids buoyant, promoting their involvement in volcanic outgassing processes and preventing sulfide sequestration (Mungall et al. 2015).
In this article, we review what controls the formation of sulfide liquids in volcanic melts, what compositions these sulfide liquids have, and the potential interaction between sulfide liquids and hydrothermal volcanic fluids. We also discuss the fundamental role that sulfides play in modulating the transport of sulfur from mantle-derived melts to the crust and into the atmosphere, and the feedbacks related to the formation of sulfide ore deposits associated with these magmas.
SULFIDE SATURATION IN ERUPTING BASALTIC MELTS
Herein, we focus on those systems that contain substantial quantities of sulfide in the melt, which correspond to magmas in ocean island settings (‘hotspots’), in mid-ocean ridge (MOR) settings, and some arc (subduction-related) magmas. The concentration of sulfur required to saturate in sulfide is known as the ‘sulfur concentration at sulfide saturation’ (SCSS). The sulfate-dominated volcanic systems are beyond the scope of this paper, but they are often characteristic of more evolved magmas and are extremely important in volcanic arcs, where anhydrite may modulate the mass of sulfur outgassing into the atmosphere (Masotta et al. 2016).
Sulfide-Saturation in Mid-Ocean Ridge Basalts (MORBs)
We will begin by considering the case of mid-ocean ridge basalts (MORBs). These basalts are well understood and characterized and are erupted with an oxygen fugacity that ensures that much of the dissolved sulfur is present as sulfide. The amount of sulfur that a melt can dissolve before saturation with respect to a sulfide phase depends on both melt composition (largely the Fe2+ content, with which sulfur forms complexes in the melt) but also temperature and pressure. Various empirical models to describe the SCSS have been proposed (Liu et al. 2007) that take account of complex compositional and intrinsic parameters. A schematic illustration to show how typical mid-ocean ridge basaltic melt may evolve due to fractional crystallization is shown in Figure 3. Sulfide saturation occurs when the sulfur concentration in the melt intersects the SCSS curve. Mid-ocean ridge basalts are thought to be sulfide-saturated during generation in their mantle source region (Mavrogenes and O’Neill 1999). The decrease in pressure during magma ascent will result in an increase in the SCSS for anhydrous melts (caused by the larger volume of dissolved sulfur over sulfide) and, thus, tend to drive the magmas toward under-saturation. Data from MORBs, however, show a correlation between sulfur concentration and decreasing Cu with MgO concentrations (Jenner and O’Neill 2012), consistent with sulfide saturation prior to and during eruption, with Cu partitioning into the sulfide phase. Observations of quenched, rounded sulfide blebs in submarine MORB glasses confirm sulfide saturation. An explanation for crustal, late-stage sulfide liquid saturation might be fractional crystallization, which even after modest amounts (~10%) will tend to drive the liquid towards sulfide saturation (Li and Ripley 2005). A detailed study of the textures of MORB sulfides show that they are typically exsolved into Ni-rich and Cu-rich regions (Fig. 4). The Ni-rich regions, known as monosulfide solid solutions (mss), are also rich in Co and Re; the Cu-rich regions, known as intermediate sulfide solid solutions (iss) are also enriched in Zn, Cd, Ag, Sn, Te, Bi and Au (Patten et al. 2013).
Basalts erupted in Iceland also display globules of quenched sulfide liquids in their groundmass (Fig. 1) (Sigmarsson et al. 2013). This suggests that, like typical MORB, they are saturated with a sulfide phase shortly before, during and after eruption, due to both fractional crystallisation (increasing the sulfur concentration in the liquid) and cooling (reducing the SCSS). Models predict that tholeiitic basaltic melts from Iceland will become saturated in a sulfide phase after around 50% crystallization, which is illustrated by data from Hekla volcano (Moune et al. 2007). Hekla melt inclusions follow a trend that predicts that they reach sulfide saturation when MgO levels in the melt reach ~6.5 wt% MgO. These levels probably result from fractional crystallization in upper crustal magma reservoirs shortly before eruption, thereby forming sulfide liquid globules as observed in erupted rocks (Fig. 1).
It is straightforward then, to demonstrate that MORB and Icelandic melts are sulfide-saturated and that sulfide saturation occurred at a relatively late stage, in response to crystallization. However, MORBs are typically erupted in water depths of > 1 km, which means that only small amounts of water are degassed from the melt and, hence, the sulfide liquids are unlikely to be in contact with aqueous magmatic fluids which in turn causes limited opportunity to transfer their sulfur burden to the atmosphere or water column (Fig. 2). But what would be the consequences of subaerial eruption and outgassing of such sulfide-bearing basalts?
Degassing of Sulfide-Saturated magmas in a Range of Tectonic Settings
As discussed above, Icelandic melts are sulfide-saturated at a relatively late stage of evolution and in response to crystallization. The very large fluxes of sulfur dioxide loading associated with Icelandic eruptions such as the 1783 Laki eruption, which produced an estimated 122 Tg of S (Thordarson and Self 1993), or the 2014 Holuhraun (Iceland) eruption, which produced an estimated 8.9 ± 0.3 Tg of S (Gauthier et al. 2016), testify to the sulfur-rich nature of the basaltic melts ascending into the crust beneath Iceland and to the efficient outgassing of sulfur from melts as they decompress and erupt. The role of sulfides in this sulfur outgassing, and its implications in terms of metal release, is worthy of further detailed consideration.
Most magmatic melts are vapour-saturated from the mid-crust, co-existing with a CO2-rich vapour phase at depth, which becomes more H2O-rich closer to the surface. The vapour phase also contains significant quantities of sulfur and halogen species. Magmatic sulfide stability is extremely sensitive to degassing-induced redox changes in the melt and to the removal, through outgassing, of exsolved sulfur species. Outgassing lowers the fugacity of sulfur in the gas phase and induces sulfide liquid (SL) oxidation and breakdown via a reaction of the form (Berlo et al. 2014):
3FeS(SL) + 2H2O(melt,fluid) + 2O2(melt,fluid) → Fe3O4 + 2H2S(fluid) + SO2(fluid)
This reaction results in the formation of magnetite Fe3O4, which does not hold metals in its structure to the same degree as sulfide. Thus, when this reaction applies, it causes the release of the concentrated metals, as well as H2S and SO2, directly into the aqueous fluid bubble–silicate melt system (Fig. 4). This process of metal transfer from the sulfide liquid to the volcanic gas phase has been inferred for a range of volcanoes, including the more evolved systems of Kawah Ijen Volcano (Berlo et al. 2014) and Merapi Volcano (Nadeau et al. 2010), both in Indonesia. At Merapi, the metal ratios observed in volcanic gases emitted from the crater have the same ratios of Cu to Au as the sulfides trapped as inclusions inside phenocrysts (Nadeau et al. 2010). At Kawah Ijen, metal concentrations in melt inclusions, combined with abundances of H2O, CO2 and S, were used to reconstruct the presence of a sulfide liquid at depth that had sequestered metals. Breakdown of the sulfide phase resulted in redistribution of metals between metals and fluids before outgassing at the volcanic vents (Berlo et al. 2014). More generally, the distribution of metals in volcanic plumes may provide corroborating evidence for such a sulfide breakdown mechanism (Fig. 5). The volcanic gas and aerosol composition of the plume accompanying the Holuhraun eruption in Iceland shows that it was enriched in metals (Gauthier et al. 2016), with their distribution mirroring the trend in elemental sulfide–silicate melt partitioning behaviour (Fig. 5). In particular, the enrichment of Re, Se and Te in the volcanic aerosol phase is strongly suggestive of an origin by sulfide breakdown in the melt because these latter elements have extremely high sulfide–silicate melt partition coefficients (>500) (Brenan 2015). Once the sulfide has broken down, there may be some further partitioning between fluid and melt, which is likely to be dependent on the chloride concentration of the fluid, owing to the tendency for metal ions to be complexed with chloride (and perhaps sulfur) in the gas phase.
In fact, the sulfide–silicate melt partition coefficients for many key metals are high (>100) (Fig. 5), and it follows that, where erupting melts are sulfide-saturated (as appears to be the case for many types of basalts), most of these metals (90.0%–99.9%) in the magma will likely be sequestered into the sulfide phase prior to eruption, even for small mass abundances of sulfide. This sulfide phase will break down when an aqueous fluid develops in the melt due to vapour saturation. The metal-rich gases observed at such volcanoes must, therefore, in large part be derived from the breakdown of sulfide liquids prior to eruption.
The sulfide liquids themselves, however, are often entirely absent in the volcanic rock products because of the efficient and rapid nature of the breakdown process described above. But in some cases, sulfide globules are preserved in the matrix glasses or, more commonly, as inclusions in phenocrysts, protected from the fluid phase by the crystal host (Fig. 1). It therefore follows that many magmas erupted at the surface have lost, due to resorption and oxidation, a large proportion of the sulfide liquids that were present in the magma prior to degassing. Additional evidence for this comes from comparisons of sulfide form and distribution within intrusive and volcanic rocks from the Cenozoic Bingham and Tintic Districts of Utah (USA). These Utah volcanic rocks (and associated porphyries) have two orders of magnitude less sulfide by mass than accompanying dykes, which were emplaced at higher pressures where degassing was inhibited. The textures of all of the sulfides in the volcanic rocks and porphyries have been modified extensively by resorption and degassing (Fig. 4). Immiscible liquids crystallized as pyrrhotite and chalcopyrite with declining temperature and pressure, and these liquids locally recrystallized to pyrite and an Fe-oxide as they were oxidized. The textures change from subspherical sulfide blebs near the margins of the quenched (and better preserved) dykes and sills, to partially altered sulfides farther in, to complete absence of sulfides in the vast majority of the intrusions and volcanic rocks (except where small sulfides are completely enclosed by phenocrysts). The cooling of these magmas, coupled with the degassing of magmatic volatiles, including sulfur-bearing gases (e.g. H2S, SO2), caused resorption and oxidation of magmatic sulfides to occur (Larocque et al. 2000). It has been estimated that degassing and oxidation removed greater than 90% of the original endowment of magmatic sulfides. This example illustrates well why volcanic rocks rarely contain sulfides in the matrix glass and provides an explanation for the low-magmatic sulfide abundances of slowly cooled, extensively degassed, large, porphyritic intrusions. Most importantly, degassing and oxidation allows metals and sulfur, under some conditions, to participate in the formation of porphyry deposits; in other cases, the metals and sulfur is outgassed to the atmosphere.
Sulfides, therefore, may play a much greater role in supplying sulfur to co-eruptive gas plumes than previously assumed. The commonly used petrological method (which uses the difference in sulfur concentration between melt inclusions and degassed matrix) to calculate sulfur outputs from past basaltic eruptions might be improved in many cases by reconstructing the sulfide saturation and producing a robust mass balance using chalcophile element inventories.
Dynamics of Sulfide–Aqueous Fluid Interaction
An important question relates to how sulfides remain in suspension in relatively low-viscosity basaltic liquids so they can participate in the resorption and oxidation reactions discussed above. Sulfide liquids are dense, and, therefore, they are predicted to settle out of the liquids by gravity and to accumulate in basal zones in the magma chamber. Such basal zones may later form the loci for economically viable Ni–Cu–platinum group element (PGE) sulfide accumulations (Ripley and Li 2013).
Recently, however, experiments have shown that the contact angle for sulfur-bearing vapour on sulfides is small (i.e. the wettability, or the ability of the fluid to maintain contact with the solid surface, is high) – much smaller than for bubbles nucleating on silicate crystal phases such as olivine, pyroxene or plagioclase (Mungall et al. 2015). Theoretical calculations also show that it is energetically far more favourable for aqueous bubbles to nucleate on sulfides or on sulfide liquid droplets, to the exclusion of all other phases, if sulfides are present, even in small amounts (Mungall et al. 2015). The effect of the formation of sulfide–bubble ‘compound drops’ on their distribution may be profound. If this process occurs at low pressures, the bulk density of the bubble–sulfide liquid aggregate may be lowered sufficiently to render the sulfide liquid buoyant relative to the silicate melt. This buoyancy allows the sulfide liquid to ascend to the surface with the melt during eruptions, participating in the outgassing process, so breaking down and supplying its sulfur and metal loads to the atmosphere (Fig. 2). The timing of vapour and sulfide saturation are potentially critical. If sulfide saturation occurs before the generation of a significant gas fraction, then sulfides may be lost and sequestered gravitationally. If vapour saturation occurs concurrently with (or before) sulfide saturation in the upper crust, then buoyant aggregates may form. The melt viscosity and the timescales of settling are also key.
THE ROLE OF SULFIDES IN OUTGASSING ASSOCIATED WITH LARGE IGNEOUS PROVINCES (LIPs)
Present-day hotspot magmatic systems in the oceans and on continents provide an analogue for ancient flood basalt provinces in terms of their sulfur budgets and the role of sulfides in outgassing. The emplacement of large igneous provinces (LIPs) have been associated with severe degradation of the Earth’s surface environment and some are coincident with extreme mass extinctions in the sedimentary record, suggesting a causal link. The potential role of sulfides in LIP degassing has implications in terms of understanding these key events in Earth history.
The sulfur budgets of LIPs are notoriously difficult to reconstruct, owing to melt inclusions being commonly small and entirely recrystallized, with rare exceptions (Self et al. 2008). It is commonly believed that LIP basalts are sulfide-saturated, a condition that forms the basis of a widely used method to calculate the likely sulfur outgassing budgets of these eruptions: using the MORB FeO–S relationship to estimate the pre-eruptive dissolved sulfur concentration and then applying the petrological method (Blake et al. 2010).
Abundant evidence of sulfide saturation and accumulation exists in LIPs worldwide, exemplified by the Norilsk sulfide-hosted PGE deposits, part of the end-Permian Siberian Traps Magmatic Province. The high Pt concentration of the deposits here require multiple episodes of sulfide resorption and precipitation to concentrate the PGE elements, as well as the assimilation of sulfate from country rocks to generate the heavy sulfur isotopic signature of the deposits (Li et al. 2009). Sediments at the Permian–Triassic boundary at many locations around the world show spikes in their metal concentrations – most notably in Ni (Rothman et al. 2014), but also other metals such as mercury (Sanei et al. 2012) – and these spikes may be linked to transport and deposition of volcanic gas/aerosols related to Siberian Traps volcanism. The direct observations of sulfide saturation in magmas supports the idea that sulfides played an important role in generating the gaseous outputs of LIPs in our geological past, perhaps contributing to these metal spikes in sediments. Further studies to understand the interplay between sulfides and sulfur/metal degassing in present-day analogues will have clear implications in terms of understanding the economic and environmental implications for LIP events in the geological record and in present-day global biogeochemical cycles and ore formation.
We acknowledge NERC urgency grant NE/M021130/1. The authors thank James Brenan, Michael Rowe and an anonymous reviewer for their helpful comments.
Berlo K, van Hinsberg VJ, Vigouroux N, Gagnon JE, Williams-Jones AE (2014) Sulfide breakdown controls metal signature in volcanic gas at Kawah Ijen volcano, Indonesia. Chemical Geology 371: 115-127
Blake S, Self S, Sharma K, Sephton S (2010) Sulfur release from the Columbia River Basalts and other flood lava eruptions constrained by a model of sulfide saturation. Earth and Planetary Science Letters 299: 328-338
Brenan JM (2015) Se–Te fractionation by sulfide–silicate melt partitioning: implications for the composition of mantle-derived magmas and their melting residues. Earth and Planetary Science Letters 422: 45-57
Carroll MR, Rutherford MJ (1985) Sulfide and sulfate saturation in hydrous silicate melts. Journal of Geophysical Research: Solid Earth 90: C601-C612
Gauthier P-J, Sigmarsson O, Gouhier M, Haddadi B, Moune S (2016) Elevated gas flux and trace metal degassing from the 2014-2015 fissure eruption at the Bárðarbunga volcanic system, Iceland. Journal of Geophysical Research: Solid Earth 121: 1610-1630
Jenner FE, O’Neill HStC (2012) Analysis of 60 elements in 616 ocean floor basaltic glasses. Geochemistry, Geophysics, Geosystems 13, doi: 10.1029/2011GC004009
Kiseeva ES, Wood BJ (2013) A simple model for chalcophile element partitioning between sulphide and silicate liquids with geochemical applications. Earth and Planetary Science Letters 383: 68-81
Larocque ACL, Stimac JA, Keith JD, Huminicki MAE (2000) Evidence for open-system behavior in immiscible Fe–S–O liquids in silicate magmas: implications for contributions of metals and sulfur to ore-forming fluids. Canadian Mineralogist 38: 1233-1249
Li C, Ripley EM (2005) Empirical equations to predict the sulfur content of mafic magmas at sulfide saturation and applications to magmatic sulfide deposits. Mineralium Deposita 40: 218-230
Li C, Ripley EM, Naldrett AJ (2009) A new genetic model for the giant Ni-Cu-PGE sulfide deposits associated with the Siberian flood basalts. Economic Geology 104: 291-301
Liu Y, Samaha N-T, Baker DR (2007) Sulfur concentration at sulfide saturation (SCSS) in magmatic silicate melts. Geochimica et Cosmochimica Acta 71: 1783-1799
Masotta M, Keppler H, Chaudhari A (2016) Fluid-melt partitioning of sulfur in differentiated arc magmas and the sulfur yield of explosive volcanic eruptions. Geochimica et Cosmochimica Acta 176: 26-43
Mather TA and 10 coauthors (2012) Halogens and trace metal emissions from the ongoing 2008 summit eruption of Kilauea volcano, Hawaii. Geochimica et Cosmochimica Acta 83: 292-323
Mavrogenes JA, O’Neill HStC (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
Moune S, Sigmarsson O, Thordarson T, Gauthier P-J (2007) Recent volatile evolution in the magmatic system of Hekla volcano, Iceland. Earth and Planetary Science Letters 255: 373-389
Mungall JE, Brenan JM, Godel B, Barnes S, Gaillard F (2015) Transport of metals and sulphur in magmas by flotation of sulphide melt on vapour bubbles. Nature Geoscience 8: 216-219
Nadeau O, Williams-Jones AE, Stix J (2010) Sulphide magma as a source of metals in arc-related magmatic hydrothermal ore fluids. Nature Geoscience 3: 501-505
Parat F, Holtz F, Streck MJ (2011) Sulfur-bearing magmatic accessory minerals. In: Behrens H, Webster JD (eds) Sulfur in Magmas and Melts: Its Importance for Natural and Technical Processes. Reviews in Mineralogy and Geochemistry 73, pp 285-314
Patten C, Barnes S-J, Mathez EA, Jenner FE (2013) Partition coefficients of chalcophile elements between sulfide and silicate melts and the early crystallization history of sulfide liquid: LA-ICP-MS analysis of MORB sulfide droplets. Chemical Geology 358: 170-188
Ripley EM, Li C (2013) Sulfide saturation in mafic magmas: Is external sulfur required for magmatic Ni-Cu-(PGE) ore genesis? Economic Geology 108: 45-58
Rothman DH and 6 coauthors (2014) Methanogenic burst in the end-Permian carbon cycle. Proceedings of the National Academy of Sciences of the United States of America 111: 5462-5467
Sanei H, Grasby SE, Beauchamp B (2012) Latest Permian mercury anomalies. Geology 40: 63-66
Self S, Blake S, Sharma K, Widdowson M, Sephton S (2008) Sulfur and chlorine in Late Cretaceous Deccan magmas and eruptive gas release. Science 319: 1654-1657
Sigmarsson O and 6 coauthors (2013) The sulfur budget of the 2011 Grímsvötn eruption, Iceland. Geophysical Research Letters 40: 6095-6100
Sigurdsson H, Devine JD, Davis A (1985) The petrologic estimation of volcanic degassing. Jökull 35: 1-8
Thordarson T, Self S (1993) The Laki (Skaftár Fires) and Grímsvötn eruptions in 1783–1785. Bulletin of Volcanology 55: 233-263
Yang X-M (2012) Sulphur solubility in felsic magmas: implications for genesis of intrusion-related gold mineralization. Geoscience Canada 39: 17-32
Zajacz Z, Candela PA, Piccoli PM, Sanchez-Valle C (2012) The partitioning of sulfur and chlorine between andesite melts and magmatic volatiles and the exchange coefficients of major cations. Geochimica et Cosmochimica Acta 89: 81-101
Zajacz Z, Halter WE, Pettke T, Guillong M (2008) Determination of fluid/melt partition coefficients by LA-ICPMS analysis of co-existing fluid and silicate melt inclusions: controls on element partitioning. Geochimica et Cosmochimica Acta 72: 2169-2197