Chromium valences in ureilite olivine and implications for ureilite petrogenesis

Abstract

Ureilites are a group of ultramafic achondrites commonly thought to be residues of partial melting on a carbon-rich asteroid. They show a large variation in FeO content (olivine Fo values ranging from ∼74 to 95) that cannot be due to igneous fractionation and suggests instead variation in oxidation state. The presence of chromite in only a few of the most ferroan (Fo 75–76) samples appears to support such a model. MicroXANES analyses were used in this study to determine the valence states of Cr (previously unknown) in olivine cores of 11 main group ureilites. The goal of this work was to use a method that is independent of Fo to determine the oxidation conditions under which ureilites formed, in order to evaluate whether the ureilite FeO-variation is correlated with oxidation state, and whether it is nebular or planetary in origin. Two of the analyzed samples, LEW 88774 (Fo 74.2) and NWA 766 (Fo 76.7) contain primary chromite; two others, LAP 03587 (Fo 74.4) and CMS 04048 (Fo 76.2) contain sub-micrometer-sized exsolutions of chromite+Ca-rich pyroxene in olivine; and one, EET 96328 (Fo 85.2) contains an unusual chromite grain of uncertain origin. No chromite has been observed in the remaining six samples (Fo 77.4–92.3).Chromium in olivine in all eleven samples was found to be dominated by the divalent species, with valences ranging from 2.10±0.02 (1σ) to 2.46±0.04. The non-chromite-bearing ureilites have the most reduced Cr, with a weighted mean valence of 2.12±0.01, i.e., Cr2+/Cr3+=7.33. All low-Fo chromite-bearing ureilites have more oxidized Cr, with valences ranging from 2.22±0.03 to 2.46±0.04. EET 96328, whose chromite grain we interpret as a late-crystallizing phase, yielded a reduced Cr valence of 2.15±0.07, similar to the non-chromite-bearing samples. Based on the measured Cr valences, magmatic (1200–1300°C) oxygen fugacities (fO2) of the non-chromite-bearing samples were estimated to be in the range IW−1.9 to IW−2.8 (assuming basaltic melt composition), consistent with fO2 values obtained by assuming olivine-silica-iron metal (OSI) equilibrium. For the primary chromite-bearing-ureilites, the corresponding fO2 were estimated (again, assuming basaltic melt composition) to be ∼IW to IW+1.0, i.e., several orders of magnitude more oxidizing than the conditions estimated for the chromite-free ureilites. In terms of Fo and Cr valence properties, ureilites appear to form two groups rather than a single “Cr-valence (or fO2) vs. Fo” trend. The chromite-bearing ureilites show little variation in Fo (∼74–76) but significant variation in Cr valence, while the non-chromite-bearing ureilites show significant variation in Fo (∼77–95) and little variation in Cr valence. These groups are unrelated to petrologic type (i.e., olivine–pigeonite, olivine–orthopyroxene, or augite-bearing). The chromite-bearing ureilites also have lower contents of Cr in olivine than most non-chromite-bearing ureilites, consistent with predictions based on Cr olivine/melt partitioning in spinel saturated vs. non-spinel-saturated systems.Under the assumption that at magmatic temperatures graphite–gas equilibria controlled fO2 at all depths on the ureilite parent body, we conclude: (1) that ureilite precursor materials having the Fo and Cr valence properties now observed in ureilites are unlikely to have been preserved during planetary processing; and (2) that the Fo and Cr valence properties now observed in ureilites are consistent with having been established by high-temperature carbon redox control over a range of depths on a plausible-sized ureilite parent body. The apparent limit on ureilite Fo values around 74–76 suggests that the precursor material(s) had bulk mg# ⩾ that of LL chondrites.