Title:Unveiling the crystallization kinetics in Ge-rich Ge$_x$Te alloys by large scale simulations with a machine-learned interatomic potential

Abstract:A machine-learned interatomic potential for Ge-rich Ge$_x$Te alloys has been developed aiming at uncovering the kinetics of phase separation and crystallization in these materials. The results are of interest for the operation of embedded phase change memories which exploits Ge-enrichment of GeSbTe alloys to raise the crystallization temperature. The potential is generated by fitting a large database of energies and forces computed within Density Functional Theory with the neural network scheme implemented in the DeePMD-kit package. The potential is highly accurate and suitable to describe the structural and dynamical properties of the liquid, amorphous and crystalline phases of the wide range of compositions from pure Ge and stoichiometric GeTe to the Ge-rich Ge$_2$Te alloy. Large scale molecular dynamics simulations revealed a crystallization mechanism which depends on temperature. At 600 K, segregation of most of Ge in excess occurs on the ns time scale followed by crystallization of nearly stoichiometric GeTe regions. At 500 K, nucleation of crystalline GeTe occurs before phase separation, followed by a slow crystal growth due to the concurrent expulsion of Ge in excess.