Physics-informed time-reversal equivariant neural network potential for magnetic materials

HY Yu and BY Liu and Y Zhong and LL Hong and JY Ji and CS Xu and XG Gong and HJ Xiang, PHYSICAL REVIEW B, 110, 104427 (2024).

DOI: 10.1103/PhysRevB.110.104427

Magnetic potential energy surface is crucial for understanding magnetic materials. This study introduces a time-reversal E(3)-equivariant neural network and physics-informed SpinGNN++ framework for constructing interatomic potentials for magnetic systems, encompassing spin-orbit coupling and noncollinear magnetic moments. SpinGNN++ integrates multitask spin equivariant neural network with explicit spin-lattice terms and time-reversal equivariant neural network to learn high-order spin-lattice interactions using time-reversal E(3)-equivariant convolutions. A complex magnetic model data set is introduced as a benchmark and employed to demonstrate its capabilities. SpinGNN++ provides accurate descriptions of the complex spin-lattice coupling in monolayer CrI(3 )and CrTe2, achieving sub-meV errors and facilitates large-scale parallel spin-lattice dynamics, thereby enabling the exploration of associated properties, including magnetic ground state and phase transition. Remarkably, SpinGNN++ identifies a differentferrimagnetic state as the ground state for monolayer CrTe2, thereby enriching its phase diagram and providing deeper insights into the distinct magnetic signals observed in various experiments.

Return to Publications page