Application of deep learning for identification of grain boundaries in atomistic simulation data

Abstract Grain boundaries (GB) characteristics, such as tilt, twist, ∑3, ∑5, etc. play a critical role in determining the microstructural and functional properties of the materials. Due to experimental challenges, atomistic simulations have become a powerful alternative for studying these GB. However, the post-processing and interpretation of simulation data often require expert input, which can introduce subjectivity into the analysis. This study proposes a deep learning (DL)-based framework for the automated segmentation of GB atoms from bulk atoms in three-dimensional atomic configurations derived from molecular dynamics (MD) simulations. We utilize a publicly available MD simulation dataset for aluminum GB structures. Raw MD simulation snapshots from this dataset are pre-processed to extract features representative of diverse GB configurations. We employ K -means clustering to explore the structural classes within the dataset, while a 3D CNN model is later used for classifying bulk and GB atoms to identify GB. The K -means clustering analysis revealed distinct patterns in the dataset for GB and bulk atoms. The 3D CNN model demonstrates classification accuracy comparable to that of the commonly used common neighbor analysis method in identifying GB atoms across a range of misorientation angles. This work highlights the potential of DL-based approaches for consistent and accurate characterization of GB in atomistic simulations.