Efficient prediction of temperature-dependent elastic and mechanical properties of 2D materials

Abstract

An efficient automated toolkit for predicting the mechanical properties of materials can accelerate new materials design and discovery; this process often involves screening large configurational space in high-throughput calculations. Herein, we present the ElasTool toolkit for these applications. In particular, we use the ElasTool to study diversity of 2D materials and heterostructures including their temperature-dependent mechanical properties, and developed a machine learning algorithm for exploring predicted properties.