An experimental assessment of the potential of sulfide saturation of the source regions of eucrites and angrites: implications for asteroidal models of core formation, late accretion and volatile element depletions

E.S. Steenstra, J. Berndt, S. Klemme, A. Rohrbach, E.S.Bullock, W. van Westrenen

Geochimica et Cosmochimica Acta
In Press, Journal Pre-proof, Available online 15 October 2019


“The geochemistry of asteroidal magmas provides fundamental clues to the processes involved in the origin and early evolution of planetary bodies. Although sulfides are important reservoirs for a diverse suite of major and trace elements, it is currently unclear whether the interiors of asteroid Vesta and the Angrite Parent Body were sulfide liquid saturated during petrogenesis of non-cumulate eucrites and volcanic angrites. To assess the potential of sulfide liquid saturation in the interiors of these bodies, high pressure (P) – temperature (T) experiments were used to quantify the sulfur concentrations at sulfide saturation (SCSS) for volcanic angrites and non-cumulate eucrites. The sulfide-silicate partitioning behavior of various trace elements was simultaneously quantified to study their geochemical behavior at sulfide liquid saturation.

It was found that the measured SCSS values agree well with the SCSS values predicted from a previous thermodynamic model for high-FeO* melts. To assess the possibility of sulfide liquid saturation of the source regions of non-cumulate eucrites and angrites, their S abundances were compared with the calculated SCSS values for their source regions. Results show that if eucritic and angritic source regions were saturated with FeS liquid, significant degassing (> 50–80%) of S must have occurred during or following their magmatic emplacement. Such loss is inconsistent with the S, Cl, Zn and Rb isotopic compositions of non-cumulate eucrites. Sulfide liquid saturation of eucrite and angrite source regions is also excluded from the strongly incompatible behavior of Cu and HSE in non-cumulate eucrites and angrites (Riches et al., 2012).

Additional calculations were performed to further explore the timing and extent of S loss during crystallization of the Vestan magma ocean. The assumption of chondritic bulk S abundances of bulk Vesta would correspond with extremely high S contents of the eucrite source region(s), even after consideration of depletion of S due to core formation. In light of the S, Cl, Zn and Rb stable isotopic compositions of eucrites, the S abundances in eucrites are most consistent with the hypothesis that the Vestan mantle was already strongly depleted in S (>70–80 %) by the time of Vestan magma ocean crystallization, resulting in more realistic S contents of the eucrite source region(s). The depletion of S could have been established during initial accretion of Vesta or it could simply reflect accretion of volatile depleted components that experienced incomplete condensation (Wu et al., 2018). Modeling of the new experimentally determined sulfide-silicate partition coefficients and previously reported Vestan mantle depletions of the various chalcophile and siderophile elements suggests that sulfide liquid segregation during early Vestan magma ocean crystallization is also unlikely. The lack of sulfide liquid saturation in the source regions of non-cumulate eucrites and angrites, as well as during early Vestan magma ocean solidification, shows that current geochemical models of core formation and late accretion remain valid for these bodies.”