A General and Scalable Parallel Hybrid Monte Carlo and Molecular Dynamics Algorithm for Alloy Simulations

Advances in metallurgy have revealed new classes of complex concentrated alloys with impressive thermomechanical, chemical, and other properties. Unraveling their underlying metallurgical mechanisms for processes like segregation and formation of precipitates that control alloy properties remains an elusive but necessary prerequisite for predicting and enhancing their performance. A powerful technique for simulating these processes is Hybrid Monte Carlo (MC)/Molecular Dynamics (MD), which combines MC steps, to rapidly alter alloy chemistry, with MD relaxations, to explore atomistic trajectories. However, traditional implementations have several performance limitations, including reliance on global energy calculations for each MC step, prohibiting the use of distributed memory parallelism. We generalize previous work by Sadigh et al. [1] to use local energy calculations for MC swaps instead and provide support in LAMMPS for popular machine learning interatomic potentials including SNAP, ACE, and FLARE. We demonstrate our generalized framework with simulations of copper zirconium interfaces and other materials and investigate opportunities for future optimizations.

[1] Sadigh, B., Erhart, P., Stukowski, A., Caro, A., Martinez, E., & Zepeda-Ruiz, L. (2012). Scalable parallel Monte Carlo algorithm for atomistic simulations of precipitation in alloys. Physical Review B—Condensed Matter and Materials Physics, 85(18), 184203.