Data-Assimilated Crystal Growth Simulation for Multiple Crystalline Phases

Abstract

To determine crystal structures from an X-ray diffraction (XRD) pattern containing multiple unknown phases, a data-assimilated crystal growth (DACG) simulation method has been developed. The XRD penalty function selectively stabilizes the structures in the experimental data, promoting their grain growth during simulated annealing. Since the XRD pattern is calculated as the Fourier transform of the pair distribution function, the DACG simulation can be performed without prior determination of the lattice parameters. We applied it to C (graphite and diamond) and SiO$_2$ (low-quartz and low-cristobalite) systems, demonstrating that the DACG simulation successfully reproduced multiple crystal structures.