Structural phase transition of two-dimensional monolayer SnTe from artificial neural network

Abstract

As machine learning becomes increasingly important in engineering and science, it is inevitable that its techniques will be applied to the investigation of materials, and in particular the structural phase transitions common in ferroelectric materials. Here, we build and train an artificial neural network to accurately predict the energy change associated with atom displacements and use the trained artificial neural network in Monte-Carlo simulations on ferroelectric materials to understand their phase transitions. We apply this approach to two dimensional single-layer SnTe and shows that it can indeed be used to simulate the phase transitions and predict the transition temperature. The artificial neural network, when viewed as a universal mathematical structure, can be readily transferred to the investigation of other ferroelectric materials when training data generated with ab initio methods are available.

Thermodynamics and dielectric response of $\text{BaTiO}_3$ by data-driven modeling

“…This framework, based on a combination of an interatomic ML potential and a vector ML model for the polarization is used to simultaneously predict the total energy, atomic forces and polarization of a ferroelectric material in order to explore its complex, temperature-dependent phase diagram as well as to predict its functional properties. The approach allows us to compute macroscopic observables -chemical potentials and dielectric susceptibilities, specifically -with an accuracy equivalent to that of the level of theory of the underlying DFT calculations; moreover, it is applicable with only minor changes to any perovskite or even any other type of ferroelectric material 27 .…”