Key points are not available for this paper at this time.
A grain boundary (GB) is a structure of great concern in materials research, which affects the mechanical properties and electrical conductivity of materials, but the microscopic thermodynamic properties of GBs cannot be explained comprehensively. In this review, we demonstrate a variety of calculation methods for GBs: density functional theory (DFT) and molecular dynamics (MDs) aim to extract the thermodynamic and kinetic properties of GBs on the atomic scale, and machine learning accelerates DFT or improves the accuracy of MDs. These methods explain the microscopic properties of a GB from different perspectives and are combined by machine learning. It is hoped that this review can inspire new ideas and provide more practical applications of computer calculations in GB engineering.
Pu et al. (Sun,) studied this question.