The Growth of Melt Inclusion- and Water-Rich Zones in Clinopyroxene Phenocrysts of the Powai Ankaramite Flow, Deccan Traps, India: Rapid Closed System Oscillatory Mineral Growth

Abstract:

Water concentrations were measured and mapped using FTIR spectroscopy in clinopyroxene phenocrysts of the Powai ankaramite flow, located near Mumbai, west of the Western Ghats escarpment of the Deccan province, India. Samples were provided by Dr. Hetu Sheth of the Indian Institute of Technology, Mumbai. Chatterjee and Sheth (2015) showed that phenocrysts in the flow were part of a cumulate layer intruded by high-temperature basaltic melt at ~ 6 kb and ~1230^oC. Cpx phenocrysts are euhedral and have concentric bands (100 to 200 microns thick) of fine (10-20 micron diameter) melt inclusions. Cpx bands that host melt inclusions have higher concentrations of water than inclusion-free bands. Water concentrations of cpx and ol were used to calculate water concentrations in the melt from which the crystals formed. Water concentrations in the parent magma were between 4.35 and 8.26 wt. % based on water concentrations in cpx, and between 8.24 and 9.41 wt. % based on those in ol. Both Mg and Fe are relatively depleted in the water- and melt inclusion-rich zones in cpx, and Ca is enriched in these zones. We suggest that oscillatory zoning in cpx is a result of repeated growth of cpx in water-richer and water-poorer boundary layers in which water lowered melt viscosity and enhanced diffusion and crystal growth rates. Water-enhanced growth rates may have resulted in preferential capture of melt inclusions preserved in water-rich cpx zones. Mg was preferentially incorporated into the cpx, causing Ca and water to build up in the boundary layer, and Mg and Fe to become relatively depleted in the boundary layer, as discussed for oscillatorially-zoned minerals by Wang and Merino (1993). Application of the equations for growth of oscillatory zones in crystals given by Wang and Merino (1993) to the growth of cpx crystals in the Powai ankaramite indicate that crystal growth occurred relatively quickly, on the order of days, although the width of the boundary zone, which is uncertain, controls the rate. These results are consistent with Wang and Merino’s contention that oscillatory growth zones in crystals can be the result of closed system processes involving advection and diffusion of ionic species into and out of the boundary zone between crystal and melt, and, on a broader scale, suggest that at least this Deccan magma was relatively hydrous.