GSA Annual Meeting in Seattle, Washington, USA - 2017

Paper No. 381-3
Presentation Time: 9:00 AM-6:30 PM

CRYSTALLIZATION OF ANORTHITIC PLAGIOCLASE: IMPACT OF MG# ON THE PSEUDO-AZEOTROPE IN THE OLIVINE-PLAGIOCLASE SYSTEM


SCHAUB, Douglas R.1, CATALANO, Tristan1, LOCKWOOD, Oliver2, NEKVASIL, Hanna1 and LINDSLEY, Donald1, (1)Dept of Geosciences, Stony Brook University, Stony Brook, NY 11794-2100, (2)Dept of Engineering, Stony Brook University, Stony Brook, NY 11794-2100, douglas.schaub@stonybrook.edu

Anorthitic plagioclase crystallization at high pressure is dictated not by the ideal melting loop configuration, but by an azeotropic topology (Lindsley, 1968), likely due to pressure-induced dissociation of the anorthite melt component as evidenced by formation of crystalline corundum and a Ca-, Si-, and albite-enriched liquid. Recent results (Nekvasil et al., 2015 GRL) demonstrate that this topology extends to lower pressures in the presence of olivine melt components and can be modeled as arising through a homogeneous melt reaction between anorthite and olivine which produces species with spinel, pyroxene, and silica stoichiometries and leads to the crystallization of aluminous spinel rather than corundum. In both the simple and olivine-bearing systems, bulk compositions to the anorthite side of the pseudo-azeotrope will exhibit anorthite enrichment of plagioclase with dropping temperature, driven by re-association of anorthite and olivine components via backreaction of the homogenous melt reaction and dissolution of corundum/spinel back into the melt. A significant unknown in these processes is the effect of Mg# on the interaction between olivine and anorthite melt components (i.e. the equilibrium constant of the homogenous melt reaction), which controls the extent of solidus depression and the Mg# of spinel species involved in this process. The latter is of particular importance for evaluating the effects of spinel dissolution on the Mg# of the melt during cooling. These factors are being assessed by a series of dynamic crystallization experiments from 0.35 to 1.0 GPa utilizing starting materials with Mg#=0.85, which will be compared with the results reported by Catalano et al. (this conference) for similar experiments with Mg#=0.5.