Fitting to magnetic forces improves the reliability of magnetic Moment Tensor Potentials
Abstract
We developed a method for fitting machine-learning interatomic potentials with magnetic degrees of freedom, namely, magnetic Moment Tensor Potentials (mMTP). The main feature of our method consists in fitting mMTP to magnetic forces (negative derivatives of energies with respect to magnetic moments) as obtained spin-polarized density functional theory calculations. We test our method on the bcc Fe-Al system with different compositions. Specifically, we calculate formation energies, equilibrium lattice parameter, and total cell magnetization. Our findings demonstrate an accurate correspondence between the values calculated with mMTP and those obtained by DFT at zero temperature. Additionally, using molecular dynamics, we estimate the finite-temperature lattice parameter and capture the cell expansion as was previously revealed in experiment. Furthermore, we demonstrate that fitting to magnetic forces increases the reliability of structure relaxation (or, equilibration), in the sense of ensuring that every relaxation run ends up with a successfully relaxed structure (the failure may otherwise be caused by falsely driving a configuration away from the region covered in the training set).
AI-Generated Overview
-
Research Focus: The development of a method for fitting magnetic Moment Tensor Potentials (mMTP) to magnetic forces alongside traditional energies, forces, and stresses to enhance the modeling of magnetic materials.
-
Methodology: The study utilized a training set derived from spin-polarized density functional theory (DFT) calculations, specifically constrained DFT (cDFT), and employed iterative optimization techniques to minimize an objective function that incorporated magnetic forces.
-
Results: The new mMTP fitting method resulted in significantly improved accuracy for the prediction of magnetic forces, achieving a tenfold reduction in error compared to previous methods. It also demonstrated 100% success in equilibrating configurations and improving overall predictive ability regarding formation energies, lattice parameters, and total magnetic moments.
-
Key Contribution(s): The paper presents a novel approach to integrate magnetic forces into the fitting procedure of mMTPs, which facilitates the accurate simulation of magnetic behaviors and enhances the reliability of structural relaxation in practical applications.
-
Significance: This research is crucial for advancing the computational modeling of magnetic materials, which are essential in numerous technologies such as spintronics and medical devices, by providing a more reliable and efficient means to capture complex magnetic interactions.
-
Broader Applications: The methodology could be applied in various fields involving magnetic materials, including the design of magnetic sensors, medical imaging devices, and materials for spintronic applications, potentially leading to advancements in technology and materials science.