Machine-learning potentials for crystal defects
Rodrigo Freitas, Yifan Cao
Abstract
Abstract Decades of advancements in strategies for the calculation of atomic interactions have culminated in a class of methods known as machine-learning interatomic potentials (MLIAPs). MLIAPs dramatically widen the spectrum of materials systems that can be simulated with high physical fidelity, including their microstructural evolution and kinetics. This framework, in conjunction with cross-scale simulations and in silico microscopy, is poised to bring a paradigm shift to the field of atomistic simulations of materials. In this prospective article we summarize recent progress in the application of MLIAPs to crystal defects. Graphical abstract
Topics & Concepts
Materials scienceFidelityMolecular dynamicsCrystal (programming language)High fidelityAtomic unitsScale (ratio)NanotechnologyComputer sciencePhysicsTelecommunicationsProgramming languageAcousticsQuantum mechanicsMachine Learning in Materials ScienceElectron and X-Ray Spectroscopy TechniquesAdvanced Electron Microscopy Techniques and Applications