Constraints on Mercury's surface composition from MESSENGER neutron spectrometer data

The composition of Mercury's surface is poorly known, but the MESSENGER (MErcury Surface, Space ENvironment, GEochemistry, and Ranging) mission has provided a wealth of new data from three flybys. In particular, MESSENGER Neutron Spectrometer (NS) observations reveal a surface enriched in neutron absorbing elements, consistent with interpretations of color and albedo observations suggesting a surface composition enriched in Fe-Mg-Ti oxides. In this study, we have computed the neutron absorption cross sections for all of the available proposed surface compositions of Mercury and evaluated the plausibility of each surface composition based on the neutron absorption cross section observed by MESSENGER. For identified plausible compositions, the implications for the thermal and magmatic evolution of Mercury are discussed. The measured macroscopic neutron absorption cross section of Mercury is inconsistent with a crust formed from partial melting of plausible bulk mantle compositions, flotation in a magma ocean or adiabatic melting of upwelling cumulates during magma ocean overturn. However, the observed neutron absorption is consistent with model compositions of late-stage magma-ocean cumulates and some proposed compositions from spectral modeling and equilibrium modeling. This suggests that the enrichment of neutron absorbing elements may be indicative of the processes that acted to form Mercury's crust. The enrichment in neutron absorbing elements, in combination with spectral observations that constrain FeO in silicates (<2 wt.%), offers strong evidence of a magma ocean on Mercury since global scale melting appears to be required to concentrate the major neutron absorbing elements while minimizing Fe in silicate minerals. We also find that iron plays a secondary role in the neutron absorption of plausible surface compositions and its variations within different Fe-Mg-Ti oxide solid solution series does not cause any overlap among the various oxide series in neutron absorption cross section. High-Fe oxides are not required and more Mg-rich oxides may even be favored as the Ti-contents can sufficiently account for the observed neutron absorption. (C) 2011 Elsevier B.V. All rights reserved.