1Bruce Fegley Jr,1Katharina Lodders,2Nathan S. Jacobson
Geochemistry (Chemie der Erde) (in Press) Link to Article [https://doi.org/10.1016/j.chemer.2023.125961]
1Planetary Chemistry Laboratory, Dept. of Earth & Planetary Sciences and McDonnell Center for the Space Sciences, Washington University, St Louis, MO 63130, USA
2Materials and Structures Division, NASA Glenn Research Center/HX5, 21000 Brookpark Road, Cleveland, OH 44135, USA
Copyright Elsevier
The chemical equilibrium distribution of 69 elements between gas and melt is modeled for bulk silicate Earth (BSE) material over a wide P – T range (1000–4500 K, 10−6–102 bar). The upper pressure end of this range may occur during lunar formation in the aftermath of a Giant Impact on the proto-Earth. The lower pressures may occur during evaporation from molten silicates on achondritic parent bodies. The virial equation of state shows silicate vapor behaves ideally in the P – T range studied. The BSE melt is modeled as a non-ideal solution and the effects of different activity coefficients and ideal solution are studied. The results presented are 50% condensation temperatures, major gas species of each element, and the pressure and temperature dependent oxygen fugacity (fO2) of dry and wet BSE material. The dry BSE model has no water because it excludes hydrogen; it also excludes the volatile elements (C, N, F, Cl, Br, I, S, Se, Te). The wet BSE model has water because it includes hydrogen; it also includes the other volatiles. Some key conclusions include the following: (1) much higher condensation temperatures in silicate vapor than in solar composition gas at the same total pressure due to the higher metallicity and higher oxygen fugacity of silicate vapor (cf. Fegley et al. 2020), (2) a different condensation sequence in silicate vapor than in solar composition gas, (3) good agreement between different activity coefficient models except for the alkali elements, which show the largest differences between models, (4) agreement, where overlap exists, with prior published silicate vapor condensation calculations (e.g., Canup et al. 2015, Lock et al. 2018, Wang et al. 2019), (5) condensation of Re, Mo, W, Ru, Os oxides instead of metals over the entire P – T range, (6) a stability field for Ni-rich metal as reported by Lock et al. (2018), (7) agreement between ideal solution (from this work and from Lock et al. 2018) and real solution condensation temperatures for elements with minor deviations from ideality in the oxide melt, (8) similar 50% condensation temperatures, within a few degrees, in the dry and wet BSE models for the major elements Al, Ca, Fe, Mg, Si, and the minor elements Co, Cr, Li, Mn, Ti, V, and (9) much lower 50% condensation temperatures for elements such as B, Cu, K, Na, Pb, Rb, which form halide, hydroxide, sulfide, selenide, telluride and oxyhalide gases. The latter results are preliminary because the solubilities and activities of volatile elements in silicate melts are not well known, but must be considered for the correct equilibrium distribution, 50% condensation temperatures and mass balance of halide (F, Cl, Br, I), hydrogen, sulfur, selenium and tellurium bearing species between silicate melt and vapor.