Equivariant Machine Learning Interatomic Potentials with Global Charge Redistribution
This work addresses limitations in computational materials science for simulating systems with long-range interactions and compositional heterogeneity, representing an incremental improvement over prior methods.
The authors tackled the problem of machine learning interatomic potentials struggling with long-range interactions and charge transfer by developing an equivariant model that incorporates global charge redistribution, which outperformed existing methods in energy and force predictions on benchmark datasets.
Machine learning interatomic potentials (MLIPs) provide a computationally efficient alternative to quantum mechanical simulations for predicting material properties. Message-passing graph neural networks, commonly used in these MLIPs, rely on local descriptor-based symmetry functions to model atomic interactions. However, such local descriptor-based approaches struggle with systems exhibiting long-range interactions, charge transfer, and compositional heterogeneity. In this work, we develop a new equivariant MLIP incorporating long-range Coulomb interactions through explicit treatment of electronic degrees of freedom, specifically global charge distribution within the system. This is achieved using a charge equilibration scheme based on predicted atomic electronegativities. We systematically evaluate our model across a range of benchmark periodic and non-periodic datasets, demonstrating that it outperforms both short-range equivariant and long-range invariant MLIPs in energy and force predictions. Our approach enables more accurate and efficient simulations of systems with long-range interactions and charge heterogeneity, expanding the applicability of MLIPs in computational materials science.