Predictive Atomistic Simulations of Materials using SNAP Data-Driven Potentials

Molecular dynamics (MD) is a powerful materials simulation method whose accuracy is limited by the interatomic potential (IAP). In many materials science applications suitably accurate potentials simply do not exist. SNAP is an automated methodology for generating application-specific IAPs using large and diverse datasets of quantum electronic structure calculations. The SNAP IAP is formulated in terms of a set of general four-body geometric invariants that characterize the local neighborhood of each atom. This approach has been used to develop potentials for diverse applications, including metal plasticity (Ta), defects in III-V semiconductors (InP), and fusion energy materials (W/Be/He/H). In each case, the SNAP IAP is fit to density functional theory calculations of energy, force, and stress for many small configurations of atoms. Cross-validation analysis and evaluation on test problems are used to further improve IAP fidelity and robustness. Varying the number of geometric descriptors allows a continuous tradeoff between computational cost and accuracy. The resultant potentials enable high-fidelity MD simulations of these materials, providing insight into their behavior on lengthscales and timescales unreachable by other methods. The relatively large per-atom computational cost of SNAP is offset by combining LAMMPS' spatial parallel algorithms with Kokkos-based hierarchical multithreading, enabling the efficient use of large CPU and GPU clusters, allocating only a few atoms to each node. Recent extensions of the SNAP approach include multi-element geometric descriptors and the use of higher-order terms.