Aubrite basalt vitrophyres: The missing basaltic component and high-sulfur silicate melts

Abstract

Aubrite basalt vitrophyres (ABVs) are clastic inclusions found in enstatite chondrites and aubrites. Three have been discovered so far; PAI from the Parsa EH3 chondrite, KTI from the Khor Temiki aubrite and L87I from the LEW 87007 aubrite. Their significance stems from the fact that their parental melts were formed from source materials similar to E chondrites; therefore, they provide an opportunity to study the process of aubrite formation from an E chondrite-like protolith. KTI, PAI and L87I contain the FeO-poor highly reduced assemblage: forsterite + enstatite + silicate glass + kamacite + troilite. Additionally, KTI and PAI contain alabandite while L87I contains diopside. ABV glass contents are: 51 vol% for KTI; 30 vol% for PAI and 13 vol% for L87I. The ABVs are, thus, representative of the different stages of crystallization of a reduced precursor basalt similar to that which gave rise to the aubrites. The chemistry of all three ABVs can be projected onto the system forsterite-albite-silica. This system is where the bulk compositions of E chondrites fall. The melting relations in this system outline the melting of aubrite parental liquids. ABV bulk compositions lie along the enstatite-forsterite reaction boundary and are generally distinct from E chondrite bulk compositions. Analysis of the forsterite-albite-silica system, and the loci of ABV bulk compositions relative to that of the enstatite chondrites shows that the ABVs can be derived by partial melting of an E chondrite protolith. The L87I vitrophyre is more oxidizing than KTI and PAI. This was determined by its: higher enstatite and forsterite FeO content, lack of alabandite, low kamacite Si content and low Ti in troilite. Forsterites enclosed in enstatite display a solid trend of anticorrelated MnO and FeO; contrary to the positive correlation found for olivines in most gecochemical settings. This anticorrelation can be explained by an oxidizing or reducing event that occurred to the L87I parental melt. L87I olivine also displays an anticorrelation of CaO with FeO. The event was determined to be an oxidizing one based upon the MnO-CaO-FeO systematics of the olivine and an analysis of the effects of increasing oxygen fugacity on the chemistry of forsterite. ABV glasses contain high amounts of dissolved S that range up to 2.5 wt% in PAI. These dissolved S contents are greater, by over an order of magnitude, than those determined by experiment under oxidizing conditions. The high S contents are the result of the S solubility mechanisms under reducing conditions where CaS° and MgS° complexes occur in the silicate melt. Both igneous and impact melt origins were considered for the ABVs. An igneous origin appears to be most consistent with: (a) ABV chemistry, petrology and texture, (b) phase equilibrium analysis of the melting and crystallization of ABV melts and (c) theoretical considerations of the melting of an E chondrite-like protolith. ABV formation can be modeled by the partial melting of an E chondrite-like protolith. The model provides a good explanation for the chemistry of ABV bulk compositions and their positions along the enstatite-forsterite reaction boundary. A partial melt origin, however, is highly inconsistent with an impact origin since impact melting in a asteroidal setting cannot produce partial melts. ABVs are good candidate materials for the “missing basaltic component” that links aubrites with their protolith material.