Partitioning of sulfur and sulfur isotopes between fluids and melts - a key for understanding and monitoring of magmatic degassing
Final Report Abstract
A new experimental approach was applied to investigate kinetics of S (and Cl) distribution between fluid and H2O-S±Cl-bearing andesitic and basaltic melt. The data provide first insights into the partitioning of S (and Cl) between fluid and melt upon degassing at disequilibrium conditions. Experiments under oxidizing conditions (> QFM+3; i.e. when S6+ is the only S species) with andesitic melt composition, revealed a strong decrease of the S content in the melt by about 85% during fast decompression (0.1 MPa/s). During subsequent annealing the released S was partly resorbed by the melt. On the other hand, at lower oxygen fugacity (QFM+1 to QFM+1.5; i.e. when S2- become abundant), a resorption of sulfur by the melt did not occur. The observations indicate a different behavior of sulfide and sulfate during kinetically-controlled degassing which need to be considered when modeling decompression induced magma degassing. In contrast to the andesitic systems, no evidence for kinetically controlled transient release of S upon fast decompression was observed for basaltic melts. The fluid-melt partitioning coefficient of sulfur, DS,fl/m, is not affected by the bulk H2O content of the melt in the range of 3 to 7 wt%, while a slight negative correlation is indicated by the experiments with andesitic composition. However, the latter trend may be due to variations in oxygen fugacity for these experiments as well as to the low accuracy of DS,fl/m derived by mass balance calculations. The addition of up to ~1000 ppm Cl to the system has a small but noticeable effect on DS,fl/m in S-enriched (>> 300 to ~3000 ppm S) andesitic systems. On the other hand, data for Cl-bearing basaltic systems shows that the S-Cl interaction is of minor importance in basaltic systems containing ≤ 0.55 wt% Cl. Noteworthy, our results show that only negligible amounts of Cl are released by Cl-bearing basaltic melts (≤ 0.36 wt% Cl) during decompression (i.e. DCl,fl/m ~ 1), while a DCl,fl/m ranging from ~1 to ~13 was determined for andesitic systems. The minor release of Cl to a fluid phase during decompression (typically from ~400 to ~70 MPa) of basaltic melts has to be considered for the interpretation of volcanic gas signatures; i.e. the Cl/S ratio in volcanic gases released by ascending basaltic magma is supposed to be very low (probably < 1) when bulk Cl is ≤ 0.55 wt%. Another important finding of our studies is that strong changes in bulk melt composition from basaltic to rhyolitic have only a minor effect on DS,fl/m under oxidizing but a major effect under reducing conditions. These differences are interpreted to be directly related to the large effect of FeO content in the melt on S2-solubility. The data obtained on S (and Cl) fluid-melt distribution at fluid-melt disequilibrium and nearequilibrium, respectively, under varying redox conditions as well as with differing initial volatile content, are of high relevance for developing volcanic degassing scenarios. SIMS measurements reveal first constraints on S-isotope fractionation between fluid and silicate melt at geologically relevant p-T condition and fluid-melt compositions. We could show that closed system degassing of S-bearing silicate melts at ~1040°C can induce S-isotope fluid-melt fractionation of about +4‰ under reducing conditions (~QFM to ~QFM+1) and of about -2 ‰ under oxidizing conditions (> QFM+3). Model calculations based on these experimental results reveal that differences in δ34S of >30 ‰ between the fluid and the source melt can be reached upon open system degassing of reduced silicate melts. Thus, monitoring S-isotopes signatures in volcanic gases with modern, high precision techniques can help to forecast volcanic events. Due to a lack of time, studies on natural samples could not be processed as planned and remain a topic for future studies, e.g. a follow-up project.
Publications
- (2014) Bubble formation during decompression of andesitic melts. American Mineralogist, 99, 1052-1062
Fiege A., Holtz F. and Cichy S.B.
(See online at https://doi.org/10.2138/am.2014.4719) - (2014) Kinetic vs. thermodynamic control of degassing of H2O-S±Cl bearing andesitic melts. Geochimica et Cosmochimica Acta, 125, 241-264
Fiege A., Behrens H., Holtz F. and Adams F.
(See online at https://doi.org/10.1016/j.gca.2013.10.012) - (2014) Sulfur Isotope Fractionation between Fluid and Andesitic Melt: An Experimental Study. Geochimica et Cosmochimica Acta, 142, 501-521
Fiege A., Holtz F., Shimizu N., Mandeville C.W., Behrens H. and Knipping J.L.
(See online at https://doi.org/10.1016/j.gca.2014.07.015) - (2015) Experimental Investigation of the S and S-Isotope Distribution between H2O-S±Cl Fluids and Basaltic Melts during Decompression. Chemical Geology, 393-394, 36- 54
Fiege A., Holtz F., Behrens H., Mandeville C.W., Shimizu N. and Crede L.S.
(See online at https://doi.org/10.1016/j.chemgeo.2014.11.012)
