Lunar rock-rain: Diverse silicate impact-vapor condensates in an Apollo-14 regolith breccia

Apollo 14 regolith breccia 14076, long known to be uniquely endowed with high-alumina, silica-poor (HASP) material of evaporation-residue origin, has been found to contain a diverse suite of complementary condensates, dubbed GASP (gas-associated spheroidal precipitates). GASP occurs in two forms: as glassy or extremely fine grained quenched-melt spheroids, mostly less than 5 lm across; and as quenched textured clasts up to 200 lm across. In two of the clasts, origin by aggregation of spheroidal GASP is confirmed by the presence of relict spheroids. GASP is distinctively depleted in the same refractory major oxides that are characteristically enriched in HASP: Al2O3 and CaO. Among the larger GASP spheroids, Al2O3 is seldom >1 wt%; among the clasts, excluding two instances of apparent contamination by Naand K-rich substrate-derived melt, bulk Al2O3 averages 0.3 wt%. Depletion of Al2O3 and CaO is also manifested by pyroxene compositions in some clasts; e.g., in the largest clast, En82Wo0.45 with 0.07 wt% Al2O3. Although GASP bulk compositions are nearly pure SiO2 + MgO + FeO, they are nonetheless highly diverse. Spheroid compositions range in mg from 7 to 84 mol%, and in FeO/SiO2 (weight ratio) from 0.002 to 0.67. Bulk compositions and textures of many GASP spheroids suggest that liquid immiscibility occurred prior to quenching; implying that these materials were, some time after condensation, at temperatures of 1680 C. Textural evidence for immiscibility includes lobate boundaries between silicic and mafic domains, and a general tendency for quenched mafic silicates to be concentrated into a few limited patches rather than evenly dispersed. The parent melt of the largest clast’s pyroxene is inferred to have formed as a partial melt within the parent aggregation of GASP matter, compositionally near the pyroxene + cristobalite + melt eutectic and thus at 1500 C. A few GASP spheroids show possible signs of in-flight collision-coalescence, but aggregation of the much larger clasts probably took place in mushy puddles on the lunar surface. Little mixing took place between these GASP puddles and the related HASP, probably because GASP condensation did not commence until after an intermediate stage during which, while neither net evaporation nor net condensation took place, expansion of the vapor cloud carried the eventual GASP matter well apart from the HASP. Considering the characteristic length-scale of lunar regolith mixing, the concentration of both GASP and HASP into this single unique regolith sample (14076) is most consistent with a parent crater size (diameter) of 10–100 km. I speculate that the 14076 regolith may have been unusually situated, almost directly uprange from an unusually oblique large impact. Mercurian analogs of the 14076 impact condensates may have significant implications for remote sensing. 2008 Elsevier Ltd. All rights reserved.