Interatomic Interaction Models for Magnetic Materials: Recent Advances

Tatiana S. Kostiuchenko,Alexander V. Shapeev,Ivan S. Novikov

2024-07-01

Abstract:Atomistic modeling is a widely employed theoretical method of computational materials science. It has found particular utility in the study of magnetic materials. Initially, magnetic empirical interatomic potentials or spin-polarized density functional theory (DFT) served as the primary models for describing interatomic interactions in atomistic simulations of magnetic systems. Furthermore, in recent years, a new class of interatomic potentials known as magnetic machine-learning interatomic potentials (magnetic MLIPs) has emerged. These MLIPs combine the computational efficiency, in terms of CPU time, of empirical potentials with the accuracy of DFT calculations. In this review, our focus lies on providing a comprehensive summary of the interatomic interaction models developed specifically for investigating magnetic materials. We also delve into the various problem classes to which these models can be applied. Finally, we offer insights into the future prospects of interatomic interaction model development for the exploration of magnetic materials.

Atomic Physics

What problem does this paper attempt to address?

This paper reviews the latest progress in atomic interaction models for magnetic materials. Traditionally, the study of magnetic materials relies on empirical potentials or density functional theory (DFT) based methods. In recent years, a new approach called machine learning interatomic potentials (magnetic MLIPs) has emerged, which combines the computational efficiency of empirical potentials and the accuracy of DFT.

The paper first introduces the importance of atomic modeling in materials science, especially for predicting the physical and mechanical properties of magnetic materials. Then, the limitations of empirical potentials such as EAM, Tersoff, and ReaxFF in describing magnetic materials are discussed. Furthermore, the paper elaborates on the accuracy of DFT in dealing with magnetic systems, but its high computational demand limits its applicability.

With the development of magnetic MLIPs, these models can simulate large systems of atoms at a lower computational cost while maintaining accuracy comparable to DFT. The paper provides an overview of different types of magnetic MLIPs, such as models based on the Heisenberg model, and how they are improved through machine learning methods to handle the complexity specific to magnetic materials.

In addition, the paper discusses the importance of magnetic materials in modern materials design, particularly in medical devices. Finally, the authors provide prospects for the future development of atomic interaction models for magnetic materials, including possible directions and challenges.

In summary, this paper aims to address how to effectively simulate and understand the complex behavior of magnetic materials, especially by developing and applying novel magnetic MLIPs to overcome the limitations of existing models.