Abstract

Plagioclase crystals in lunar anorthosite may have formed directly from the lunar magma ocean (LMO), with their elemental and isotopic compositions providing unique probes of LMO composition, age, and evolution. Low hydrogen concentrations have been reported in lunar plagioclase, but translating this to estimates of LMO hydrogen contents is complicated by incomplete knowledge of hydrogen partitioning between plagioclase and melt at low hydrogen contents. We conducted a series of high-temperature experiments to better quantify the plagioclase-melt partition coefficient of hydrogen in the Moon at low melt H abundances. Hydrogen contents of plagioclase and coexisting melt, reported in terms of water equivalent concentrations, were analyzed using Fourier Transform Infrared Spectroscopy. Resulting plagioclase-melt partition coefficients of water, Dplag-melt water, at melt water contents below 300 ppm, are 0.10 ± 0.05 to 0.36 ± 0.19. the highest Dplag-melt water reported to date. Our results, in conjunction with literature data, indicate a strong dependence of Dplag-melt water on water concentration in the melt (x in ppm): Dplag-melt water=20±5·x−0.88±0.03. This strong non-Henrian behavior is likely related to changes in the hydrogen activity in basaltic melt due to variations in H2O/OH speciation as a function of melt hydrogen content, implying that hydrogen partitioning in other mineral-melt systems may also be non-Henrian. At the low water contents in the lunar interior, Dplag-melt water is approximately 2 orders of magnitude higher than previously assumed. If hydrogen contents measured in lunar plagioclase crystals reflect pristine plagioclase-LMO equilibrium without subsequent modification, we conclude that <14 ppm H2O equivalent would have been present in the Moon when 95% of the initial LMO had crystallized.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call