A Machine-Learned Spin-Lattice Potential for Dynamic Simulations of Defective Magnetic Iron

TL;DR

This paper introduces a machine-learned spin-lattice potential for magnetic iron, enabling accurate and stable simulations of defects and magnetic behavior at mesoscopic scales, surpassing previous models.

Contribution

The development of a neural network-augmented spin-lattice potential for magnetic iron that accurately models defects and magnetic properties at larger scales.

Findings

Reproduces cohesive energies of BCC and FCC phases.

Accurately predicts defect formation energies and magnetic structures.

Demonstrates stability at high temperatures in spin-lattice dynamics.

Abstract

A machine-learned spin-lattice interatomic potential (MSLP) for magnetic iron is developed and applied to mesoscopic scale defects. It is achieved by augmenting a spin-lattice Hamiltonian with a neural network term trained to descriptors representing a mix of local atomic configuration and magnetic environments. It reproduces the cohesive energy of BCC and FCC phases with various magnetic states. It predicts the formation energy and complex magnetic structure of point defects in quantitative agreement with density functional theory (DFT) including the reversal and quenching of magnetic moments near the core of defects. The Curie temperature is calculated through spin-lattice dynamics showing good computational stability at high temperature. The potential is applied to study magnetic fluctuations near sizable dislocation loops. The MSLP transcends current treatments using DFT and molecular dynamics, and surpasses other spin-lattice potentials that only treat near-perfect crystal cases.