ATOMIC-scale foundation of a Physics-based Interface Engineering in crystalline materials

Inorganic materials are integral to numerous fields, including transportation, electronics, and medical applications. Understanding their mechanical properties, particularly permanent inelastic deformation, has been a central pursuit in material science. My project, AtomicPIE, focuses on uncovering the key mechanisms driving plasticity in metallic alloys, emphasizing dislocation-interface interactions. These interactions occur at grain boundaries and phase boundaries, which significantly influence material strength, reliability, and lightweight properties. Despite extensive studies, a comprehensive understanding of slip transfer—how dislocations propagate across interfaces—remains elusive. AtomicPIE integrates atomistic simulations, generative machine learning models, and enhanced discrete dislocation dynamics to address this gap. By simulating crystalline defects and their interactions, the project employs Nye dislocation densities to bridge atomic- to micro-scale modeling. A key innovation is the use of machine learning to predict complex Nye dislocation fields for mesoscale simulations. The framework developed within AtomicPIE will be validated through experiments and cross-scale molecular dynamics, focusing on ultra-fine-grain aluminum alloys. AtomicPIE aims to establish a physics-based understanding of slip transfer at interfaces, paving the way for advanced materials design and interface engineering.