¹D. P. Moriarty III, ¹C. M. Pieters
¹Department of Earth, Environmental and Planetary Sciences, Brown University, Providence, Rhode Island, USA

We reexamine the relationship between pyroxene composition and near-infrared absorption bands, integrating measurements of diverse natural and synthetic samples. We test an algorithm (PLC) involving a two-part linear continuum removal and parabolic fits to the 1 and 2 μm bands—a computationally simple approach which can easily be automated and applied to remote sensing data. Employing a suite of synthetic pure pyroxenes, the PLC technique is shown to derive similar band centers to the modified Gaussian model. PLC analyses are extended to natural pyroxene-bearing materials, including (1) bulk lunar basalts and pyroxene separates, (2) diverse lunar soils, and (3) HED meteorites. For natural pyroxenes, the relationship between composition and absorption band center differs from that of synthetic pyroxenes. These differences arise from complexities inherent in natural materials such as exsolution, zoning, mixing, and space weathering. For these reasons, band center measurements of natural pyroxene-bearing materials are compositionally nonunique and could represent three distinct scenarios (1) pyroxene with a narrow compositional range, (2) complexly zoned pyroxene grains, or (3) a mixture of multiple pyroxene (or nonpyroxene) components. Therefore, a universal quantitative relationship between band centers and pyroxene composition cannot be uniquely derived for natural pyroxene-bearing materials without additional geologic context. Nevertheless, useful relative relationships between composition and band center persist in most cases. These relationships are used to interpret M3 data from the Humboldtianum Basin. Four distinct compositional units are identified (1) Mare Humboldtianum basalts, (2) distinct outer basalts, (3) low-Ca pyroxene-bearing materials, and (4) feldspathic materials.

References
Moriarty III DP, Pieters CM (2016) Complexities in pyroxene compositions derived from absorption band centers: Examples from Apollo samples, HED meteorites, synthetic pure pyroxenes, and remote sensing data. Meteoritics&Planetary Science (in Press)
Link to Article [DOI: 10.1111/maps.12588]
Published by arrangement with John Wiley&Sons

¹E.S. Steenstra, ¹J.S. Knibbe, ^2,3N. Rai, ¹W. van Westrenen
¹Faculty of Earth and Life Sciences, VU University Amsterdam, Amsterdam, The Netherlands
²Centre for Planetary Sciences, Birkbeck – UCL, London, UK
³Department of Earth Sciences, Mineral and Planetary Sciences Division, Natural History Museum, London, UK

It is now widely accepted that the asteroid 4–Vesta has an Fe–rich metallic core, but the composition of the core and the conditions prevailing during core–mantle differentiation are poorly constrained. In light of new constraints on Vesta’s geophysical and geochemical properties obtained by the DAWN mission, we have re–examined the conditions at which core–mantle differentiation in Vesta may have occurred by linking the estimated mantle depletions of siderophile elements P, Co, Ni, Cu, Ga, Ge, Mo and W in the vestan mantle to newly derived predictive equations for core–mantle partitioning of these elements. We extend the number of elements previously considered in geochemical modeling of vestan core formation and use published metal–silicate partitioning data obtained at low pressures to characterize the dependence of metal/silicate partition coefficients (D) on pressure, temperature, oxygen fugacity and composition of the silicate and metallic melt. In our modeling we implement newly derived mantle depletions of P, Co, Ni and Ga through analysis of published HED meteorite analyses and assess two contrasting bulk compositional models for Vesta.

Modeling results using Monte Carlo simulations constrain vestan core formation to have occurred at mildly reducing conditions of approximately 2 log units below the iron–wüstite (IW) buffer (ΔIW = –2.05±0.20) if the two most likely bulk compositions (binary mixtures of H + CM or H + CV chondritic meteorites) are considered, assuming a temperature range between 1725–1850 K and a sulfur–free pure Fe core. If the core is assumed to be sulfur–rich (15 wt.% S) as predicted by the latter bulk compositional models, observed depletions for all eight siderophile elements can be simultaneously satisfied at ΔIW = –2.35±0.10 and 1725–1850 K for the H + CV bulk composition and ΔIW = –2.30±0.15 and 1725–1850 K for the H + CM bulk composition. More reducing conditions are not consistent with the observed siderophile element depletions in the vestan mantle, independent of the sulfur content of the vestan core and of the bulk compositional model chosen.

Our analysis shows that a previously proposed shallow magma ocean on Vesta during core formation is not consistent with the observed mantle depletion of Ga for the two considered bulk compositions irrespective of core composition. Instead, our results are consistent with the existence of a deep magma ocean during core formation on Vesta, requiring >50 to 100 per cent mantle melting.

References
Steenstra ES, Knibbe JS, Rai N, van Westrenen W (2016) Constraints on core formation in Vesta from metal–silicate partitioning of siderophile elements. Geochimica et Cosmochimica Acta (in Press)
Link to Article [doi:10.1016/j.gca.2016.01.002]
Copyright Elsevier