Estimation of the diffusion coefficient of GaSb in InSb melt using Bayesian optimization and the ISS experimental results

Abstract The diffusion coefficient of GaSb in InSb melt was estimated by using the microgravity experimental results performed onboard the International Space Station (ISS), where the effect of natural convection is minimized. We numerically investigated the transport phenomena during InGaSb crystals growth via a vertical gradient freezing method and it was deduced that the growth rate is dominated by the solute diffusion in the melt. Furthermore, the crystal growth rate was well explained by adopting a diffusion coefficient with function of GaSb concentration, obtained by analyzing the available experimental results and utilizing machine learning and Bayesian optimization methods. These results also contribute to further understanding of crystal growth on the ground.