Supersaturation Nucleation and Growth of Plagioclase: a numerical model of decompression-induced crystallization

Abstract

Supersaturation Nucleation and Growth of Plagioclase (SNGPlag) is a numerical model that predicts the nucleation and growth of plagioclase crystals in a decompressing magma as a function of time. The model is written in Matlab, but is available as a standalone compiled program. SNGPlag uses the MELTS webservice to determine equilibrium plagioclase mode, for a user-defined magma composition, as a function of pressure and temperature. User inputs include decompression path, the presence and size distributions of antecrysts and phenocrysts, and crystal shape. At each time step, the model evaluates the difference between the calculated crystallinity and equilibrium crystallinity for a given pressure and temperature to determine the degree of supersaturation, which then sets plagioclase nucleation and growth rates. Growth rates are used to grow the existing crystals whereas nucleation adds new crystals. SNGPlag produces results that can be compared to quantitative textures in natural volcanic rocks, including total crystallinity, microlite number density, microlite crystal size distribution, the characteristic size of microlite crystals, as well as a time series of crystallinity. Model results are consistent with the established crystallization theory. As expected, microlite crystallinity increases as decompression rate slows. Decompression path greatly affects microlite textures. For the same average decompression rate, single-step paths have higher crystallinities and microlite number densities than multi-step decompressions, which are in turn more crystalline than continuous paths. Pre-existing crystals damp microlite crystallization, as these crystals provide a substrate to accommodate crystal growth and thus reduce supersaturation. The size distribution and volume fraction of these pre-existing crystals determines the magnitude of the damping. SNGPlag predicts that melt composition and temperature also exert important controls. Higher temperatures and higher silica contents both reduce microlite crystallization. In comparison with the previous studies of decompression rate based on microlite crystallization experiments, SNGPlag generally predicts minimum decompression rates that are up to three-to-four times slower. The difference is likely because those studies applied single- or multi-step decompression experiments to simulate natural magma ascent, which may be better represented by continuous decompression pathways or series of continuous decompression intervals punctuated with pauses. Previous studies also fail to account for the effects of phenocrysts or antecrysts on microlite nucleation and growth.