HR: 0800h
AN: V41A-1430    [Abstracts]
TI: Can Sulfur Speciation in Olivine-Hosted Melt Inclusions be Accurately Determined by Electron Microprobe: An Experimental Study with a Practical Application
AU: * Rowe, M C
EM: rowem@geo.oregonstate.edu
AF: Oregon State University, Department of Geosciences, Corvallis, OR 97331 United States
AU: Kent, A J
EM: adam.kent@science.oregonstate.edu
AF: Oregon State University, Department of Geosciences, Corvallis, OR 97331 United States
AU: Nielsen, R L
EM: nielsenr@geo.oregonstate.edu
AF: Oregon State University, Department of Geosciences, Corvallis, OR 97331 United States
AU: Wallace, P J
EM: pwallace@darkwing.uoregon.edu
AF: University of Oregon, Department of Geological Sciences, Eugene, OR 97403 United States
AU: Donovan, J
EM: donovan@darwing.uoregon.edu
AF: University of Oregon, Department of Geological Sciences, Eugene, OR 97403 United States
AB: Determination of sulfur speciation, based on the S Kα peak shift, by electron microprobe analysis has been well documented for glasses, however analysis of S speciation in melt inclusions provides additional complications. Apart from issues related to smaller sample sizes, rehomogenization of melt inclusions prior to analysis also has the potential to alter redox conditions. To study the applicability of this technique to olivine-hosted melt inclusions, we have conducted studies of S speciation for homogenized melt inclusions heated over a range of temperatures and durations from submarine and subaerial lavas. Overall there is an apparent reduction of the S in heated melt inclusions that increases with longer heating times, producing a shift in the S Kα peak position of up to 0.86*10^-3 Å relative to the S peak position of the matrix glass and unheated inclusions. However, for short heating durations (< 10min) this shift is minimal (< 0.2*10^-3 Å). In subaerially erupted lavas there is an additional complication in that in general, the mean oxidation state of S in volcanic gas is not the same as in the degassing magma, so the S⁶⁺/S^2- in residual melt may change during degassing. However, in submarine and rare subaerial lavas, S speciation in naturally glassy melt inclusions is consistent with matrix glass from the same sample. As with analysis of glasses, exposure to the electron beam has also been shown to oxidize the sulfur present in glass inclusions during analysis. However, the degree to which this affects the S speciation is dependant on the S concentration and initial S⁶⁺/S^2- of the glass. In order to avoid this complication, the electron beam is incrementally moved (1 micron/min) over the melt inclusion. Studies of melt inclusions in basalts from Cascadia show decreasing S⁶⁺/S^2- with increasing distance from the trench, consistent with chromite-olivine oxygen barometery and indicative of a general decrease in oxidation state. A similar trend is observed in Cl concentrations with the highest concentrations in the forearc (1800ppm) decreasing to the backarc (< 700ppm). The general correlation of S speciation and concentration and Cl concentration suggests that these volatile elements are likely derived, at least in part, from the subducting Juan de Fuca plate.
DE: 1009 Geochemical modeling (3610, 8410)
DE: 1031 Subduction zone processes (3060, 3613, 8170, 8413)
DE: 1043 Fluid and melt inclusion geochemistry
DE: 1094 Instruments and techniques
DE: 3630 Experimental mineralogy and petrology
SC: Volcanology, Geochemistry, Petrology [V]
MN: Fall Meeting 2005