Sulphur behaviour and redox conditions in etnean magmas during magma differentiation and degassing

Abstract

Sulphur behaviour and variations in redox conditions during magma differentiation and degassing in Mt. Etna (Italy) volcanic system have been explored by integrating the study of olivine-hosted melt inclusions (MIs) with an experimental survey of sulphur solubility in hydrous basaltic magmas. Sulphur solubility experiments were performed at conditions relevant to the Etnean plumbing system (1200 °C, 200 MPa and oxygen fugacity between NNO + 0.2 and NNO + 1.7, with NNO being the Nickel-Nickel Oxide buffer), and their results confirm the important control of oxygen fugacity (fO2) on S abundance in mafic magmas and on S partitioning between fluid and melt phases (DSfluid/melt). The observed DSfluid/melt value increases from 51 ± 4 to 146 ± 6 when fO2 decreases from NNO + 1.7 ± 0.5 to NNO + 0.3. Based on the calculated DSfluid/melt and a careful selection of previously published data, an empirical model is proposed for basaltic magmas in order to predict the variation of DSfluid/melt values upon variations in P (25–300 MPa), T (1030–1200 °C) and fO2 (between NNO-0.8 and NNO + 2.4). Olivine-hosted melt inclusions (Fo89-91) from tephra of the prehistoric (4 ka BP) sub-plinian picritic eruption, named FS (“Fall Stratified”), have been investigated for their major element compositions, volatile contents and iron speciation (expressed as Fe3+/ƩFe ratio). These primitive MIs present S content from 235 ± 77 to 3445 ± 168 ppm, while oxygen fugacity values, estimated from Fe3+/ƩFe ratios, range from NNO + 0.7 ± 0.2 to NNO + 1.6 ± 0.2. Iron speciation has also been investigated in more evolved and volatile-poorer Etnean MIs. The only primitive melt inclusion from Mt. Spagnolo eruption (4–15 ka BP) presents a S content of 1515 ± 49 ppm and an estimated fO2 of NNO + 1.4 ± 0.1. The more evolved MIs (from 2002/2003, 2006, 2008/2009 and 2013 eruptions) have S content lower than 500 ppm, and their Fe3+/ΣFe ratios result in fO2 between NNO-0.9 ± 0.1 and NNO + 0.4 ± 0.1. Redox conditions and S behaviour in Etnean magmas during degassing and fractional crystallization were modelled coupling MELTS code with our empirical DSfluid/melt model. Starting from a FS-type magma composition and upon decrease of T and P, fractional crystallization of olivine, clinopyroxene, spinel and plagioclase causes a significant fO2 decrease. The fO2 reduction, in turn, causes a decrease in sulphur solubility and an increase in DSfluid/melt, promoting S exsolution during magma ascent, which further enhances the reduction of fO2. For the evolved melt inclusions of 2002–2013 eruptions, magma differentiation may therefore have played a crucial role in decreasing redox conditions and favouring efficient S degassing. Differently, during the unusual FS eruption, only limited melt evolution is observed and S exsolution seems to have been triggered by a major pressure decrease accompanied by H2O and CO2 exsolution during fast magmatic ascent.

Storage, evolution, and mixing in basaltic eruptions from around the Okataina Volcanic Centre, Taupō Volcanic Zone, Aotearoa New Zealand

“…The low Cl and ST concentrations indicates partitioning into a coexisting fluid. This is expected at low pressures and the hot, dry melt conditions observed; especially for sulphur in more evolved melt compositions (e.g., Clemente et al, 2004;Gennaro et al, 2021;O'Neill, 2020;Tattitch et al, 2021;Thomas and Wood, 2021) (Figure 4). Dacite-Tw is chemically distinct from Dacite-Rm and melt inclusions record a lower temperature of ~850 °C (pressures could not be estimated from the available data, Figure 5a).…”

“…Concentrations of ST and Cl increase in the melt during crystallisation for melt inclusions with >1000 ppm ST, indicating these volatiles behave incompatibly (i.e., not partitioned into coexisting solids or exsolved fluids) (Figure 4). The magma may have been volatile-undersaturated, and therefore there was no fluid phase for Cl or S to partition into, or fluid-melt partition coefficients for S and Cl were low (e.g., Gennaro et al, 2021;O'Neill, 2020;Tattitch et al, 2021;Thomas and Wood, 2021). If the magma was initially fluid-undersaturated, this would contrast with most arc regions where high magmatic CO2 concentrations result in vapour-saturation deep in the crust (e.g., Wallace, 2005).…”

Sulfur_X: A model of sulfur degassing during magma ascent

“…All these forward models are based on parameters relating the S concentration or fugacity in the gas and the S concentration in the silicate melt, such as sulfide/sulfate solubilities (Burgisser et al, 2015), sulfide/sulfate capacities (Scaillet and Pichavant, 2003;Moretti and Papale, 2004), and sulfur partition coefficients (Witham et al, 2012). However, none of these parameters were well calibrated by xperiments at pressures higher than 300 MPa (O'Neill and Mavrogenes, 2002;Moune et al, 2008;Lesne et al, 2011;Zajacz et al, 2012;Zajacz et al, 2013;Fiege et al, 2014b;Zajacz, 2014;Beermann et al, 2015;Fiege et al, 2015;Masotta et al, 2016;Nash et al, 2019;Gennaro et al, 2020), which might have led to inaccurate extrapolation at high pressures.…”

“…Furthermore, some of the existing experiments could suffer from the formation of sulfide solids (Masotta et al, 2016;Gennaro et al, 2020), inaccurate estimate of fO2 (Lesne et al, 2011;Fiege et al, 2014b;Fiege et al, 2015), S loss to the capsule (Lesne et al, 2011), which might have caused significant discrepancies among existing degassing models even at pressures of 300 MPa or lower.…”

Sulphur behaviour and oxygen fugacity variation in Mt. Etna system revealed by melt inclusions

“…The picrobasaltic MIs are extremely enriched in H 2 O and CO 2 (up to 5.9 and 0.58 wt.% respectively). S content in Etnean MIs ranging from some ppm up to more than 4000 ppm, whereas the oxygen fugacity (fO 2 , estimated using XANES Fe 3+ /ƩFe ratios and oxybarometers models) ranges from NNO-0.9±0.1 to NNO+1.6±0.2, with NNO being the Nickel-Nickel Oxide buffer [1].…”