Dynamic crystallization of shock melts in Allan Hills 77005: Implications for melt pocket formation in Martian meteorites

Abstract

A series of crystallization experiments have been performed on synthetic glasses matching the composition of a melt pocket found in Allan Hills (ALH) 77005 in order to evaluate the heterogeneous nucleation potential of the melt and the effect of oxygen fugacity on crystallization. The starting temperature of the experiments varied from superliquidus, liquidus and subliquidus temperatures. Each run was then cooled at rates of 10, 500 and 1000 °C/h at FMQ. The results of this study constrain the heating and cooling regime for a microporphyritic melt pocket. Within the melt pocket, strong thermal gradients existed at the onset of crystallization, giving rise to a heterogeneous distribution of nucleation sites resulting in gradational textures of olivine and chromite. Skeletal olivine in the melt pocket center crystallized from a melt containing few nuclei cooled at a fast rate. Nearer to the melt pocket margin elongate skeletal shapes progress to hopper and equant euhedral, reflecting an increase in nuclei in the melt at the initiation of crystallization and growth at low degrees of supercooling. Cooling from post-shock temperatures took place on the order of minutes. An additional series of experiments were conducted for a melt temperature of 1510 °C and a cooling rate of 500 °C/h at the FMQ buffer, as well as 1 and 2 log units above and below FMQ. Variation in chromite stability in these experiments is reflected in crystal shapes and composition, and place constraints on the oxygen fugacity of crystallization of the melt pocket. We conclude that the oxygen fugacity of the melt pocket was set by the Fe 3+/Fe 2+ ratio imparted by melting of the host rock, rather than external factors such as incorporation of CO 2-rich Martian atmosphere, or melting and injection of oxidized surface (e.g., regolith) material. Comparison with previous crystallization experiments on melt pockets in Martian basalts indicate that the predominance of dendritic crystal shapes reflects the likelihood that those melt pockets with lower liquidus temperatures will be more completely melted, destroying most or all nuclei in the melt. In this case, crystal growth takes place at high degrees of supercooling, yielding dendritic shapes. It appears as though the melting process is just as important as cooling rate in determining the final texture of the melt pocket, as this process controls elimination or preservation of nuclei at the onset of cooling and crystallization.