Automated identification of stacking faults and twin boundaries in face-centered cubic crystal

The primary objective of atomic-scale simulations is to meticulously monitor the formation and evolution of microstructures and defects, and quantitatively analyze their effects on mechanical and physical properties. In face-centered cubic (FCC) metals, planar defects such as stacking faults (SFs), twin boundaries (TBs), and local hexagonal close-packed (HCP) structures have a significant influence on mechanical properties. However, the existing methods for identifying microstructures classify SF, TB, and local HCP phase transitions as HCP structures, without further differentiation. To address this issue, an approach is proposed that can further identify the defect structures by meticulously examining the local structural environment and neighbors of atoms. Based on a density-based clustering algorithm in the particle-position datasets, this method can not only clearly distinguish these planar defects but also provide more quantitative information about them, such as their quantity, area, and number of atoms involved, which is crucial for establishing a quantitative relationship between different planar defects and their contents within FCC structural materials, and their mechanical properties.