Zr Diffusion in Rhyolitic Melt During Zircon Dissolution

Zr diffusion rate in silicate melt is one major controlling factor on how long zircon can survive in magma or how fast zircon can grow. Even though four papers investigated Zr diffusion in dry and wet rhyolitic melts (Harrison and Watson, 1983; Baker and Watson, 1988; Baker 1990, 2002), the data are not consistent. For example, for dry rhyolitic melt, data of Harrison and Watson (1983), Baker and Watson (1988) and Baker (1990, 2002) can differ by a factor of 25. For hydrous rhyolitic melt, Zr diffusivities in rhyolitic melt with 6.0 wt% H2O (Harrison and Watson, 1983) are smaller than those with 4.2-4.8 wt% H2O (Baker, 2002). Some of the early data might be compromised by convection during the experiments, or by inaccurate H2O concentration determination. We have carried out new zircon dissolution experiments, paying special attention to avoid convection and characterizing H2O content both before and after the experiment. Highly reproducible Zr diffusivity data are obtained on a dry rhyolitic melt with 0.12 wt% H2O and a wet rhyolitic melt with 5.5 wt% H2O (post-experimental analyses show 5.2-5.9 wt% H2O). Our new data on dry rhyolitic melt lie between data of Harrison and Watson (1983) and those of Baker and Watson (1988) and Baker (1990, 2002), but closer to Harrison and Watson (1983). Furthermore, our new data show that Zr diffusivity does not change significantly as SiO2 content changes from 72.8 to 76.3 wt%. On hydrous rhyolitic melt, our data for a rhyolite with 5.2-5.9 wt% H2O are higher than those of Harrison and Watson (1983) on a rhyolite with 6 wt% H2O and lower than those of Baker (2002) on a rhyolite with 4.3-4.8 wt% H2O, but close to the latter. More work will be carried out so that Zr diffusivity in rhyolitic melt can be predicted as a function of temperature, pressure and H2O content.