Are All Obsidians Super-Heated? Insights from Observations of Crystallization Kinetics in Experiments on Glass Mountain Obsidians (Long Valley, CA)

Monday, 14 December 2015: 08:30
104 (Moscone South)
Laura Waters, Smithsonian Institution, National Museum of Natural History, Washington, DC, United States and Benjamin James Andrews, Smithsonian Institution, Washington, DC, United States
The Glass Mountain obsidians (Long Valley, CA) are crystal-poor (<8%) and highly-evolved (high SiO2, low MgO, Sr, Ba) and, therefore, their formation required extremely efficient crystal-liquid separation. Petrologic and experimental investigation of the mineral phases in Glass Mountain lavas may reveal differentiation processes that generated the obsidians, if the mineral assemblage is phenocrystic. Results of high-resolution SEM mapping and electron microprobe analysis of a Glass Mountain sample reveal that the obsidian is saturated in nine phases (sanidine + quartz + plagioclase + ilmenite + titanomagnetite + zircon + apatite + allanite + biotite). Sanidine (Or78-Or35) and quartz occur in the largest abundances, and plagioclase (<An20) is the third most abundant. Sanidine crystals have a “granophyric” texture, marked by intergrowths of quartz and sanidine. H2O-saturated phase equilibrium experiments conducted in cold-seal pressure vessels conducted over a range of conditions (700-850°C; 75-225MPa) show that mineral phases in the natural sample are plausible phenocrysts with compositions that record degassing ± cooling. A surprising result from the phase equilibrium experiments is that, though phenocrysts from the natural sample reflect equilibrium compositions, they do not reflect equilibrium abundances. The low crystallinity, saturation in multiple phases, “granophyric” texture in sanidine, and glassy nature of the obsidians, requires that the mechanism that produced these obsidians have an associated kinetic effect that strongly hinders nucleation. Decompression and cooling experiments, conducted in this study and from the literature, demonstrate that the simplest way to hinder nucleation is to initiate degassing or cooling from super-liquidus conditions. Therefore, the Glass Mountain obsidians were super-heated prior to crystallization, achieved either by fluid under-saturated decompression from a crystalline mush or H2O-saturated partial melting.