Effects of pH_2O, pH_2 and fO_2 on the diffusion of h-bearing species in lunar basaltic liquid and an iron-free basaltic analog at 1 atm

Abstract

We have experimentally determined the diffusivity of water in a representative lunar basaltic liquid composition (LG) and in an iron-free analog of a basaltic liquid (AD) at the low water concentrations and low oxygen fugacities (fO_2) relevant to the eruption of lunar basalts. Experiments were conducted at 1 atm and 1350 °C over a range of pH_2/pH_2O from near zero to ∼10 and a range in fO_2 spanning ∼9 orders of magnitude (from 2.2 log units below the iron-wüstite buffer, IW–2.2, to IW+6.7). The water concentrations measured in our quenched experimental glasses by secondary ion mass spectrometry (SIMS) and Fourier transform infrared spectroscopy (FTIR) vary from a few ppm to ∼430 ppm. Water concentration gradients in the majority of our AD experiments are well described by models in which the diffusivity of water (D*_(water)) has a constant value of ∼2×10^(–10) m^2/s, while our LG results indicate that D*_(water) in LG melt has a constant value of ∼6×10^(–10) m^2/s under the conditions of our experiments. Water concentration gradients in hydration and dehydration experiments that were run simultaneously in H_2/CO_2 gas mixtures are well described by the same D*_(water), and water concentrations measured near the melt-vapor interfaces of these experiment pairs are approximately the same. These observations strongly support an equilibrium boundary condition for our experiments containing >70 ppm H_2O. However, dehydration experiments into nominally anhydrous CO_2, N_2, and CO/CO_2 gas mixtures leave some scope for the importance of kinetics during dehydration of melts containing less than a few 10's of ppm H_2O. Comparison of our results with the modified speciation model (Ni et al., 2013) in which both molecular water and hydroxyl are allowed to diffuse suggests that we have resolved the diffusivity of hydroxyl (D_(OH)) in AD and LG melts. Our results support a positive correlation between D_(OH) and melt depolymerization. Best-fit values of D*_(water) for our LG experiments vary within a factor of ∼2 over a range of pH_2/pH_2O from 0.007 to 9.7 and a range of logf O_2 from IW–2.2 to IW+4.9. The relative insensitivity of our best-fit values of D*_(water) to variations in pH_2 suggests that H_2 diffusion did not control the rate of degassing of H-bearing species from the lunar glasses of Saal et al. (2008); however, we cannot rule out a role for molecular H2 diffusion under lower-temperature and/or higher-pressure conditions than explored in our experiments. The value of D*_(water) chosen by Saal et al. (2008) for modeling the diffusive degassing of the lunar volcanic glasses is within a factor of ∼2 of our measured value in LG melt at 1350 °C. By coupling our LG results at 1350 °C with an activation energy of 220 kJ/mol (Zhang et al. 2017), we obtain the following Arrhenius relationship, which can be used to model syneruptive diffusive water loss from lunar melt beads: D*_(water)(m^2/s) = 7.2×10^(-3)exp(-2.6×10^4/T(K)).

Additional details