Interatomic Potential Prediction Based on Charge Equilibration and Equivariant Transformer
Main Article Content
Keywords
interatomic potential, equivariant Transformer, graph neural network, machine learning
Abstract
Conventional local machine-learning interatomic potentials describe atomic environments with a finite cutoff radius and therefore have difficulty capturing long-range electrostatic coupling in polar, charged, or charge-transfer systems. This paper proposes CE-ETNet (Charge-Equilibration-Enhanced Equivariant Transformer Network), an equivariant Transformer interatomic potential constrained by charge equilibration. The model learns local chemical environments formed by atom types, geometric edges, and radial basis features through an equivariant representation encoder, and predicts atomic electronegativities. Under the constraint of total charge conservation, a differentiable charge equilibration procedure is used to solve partial charges and electrostatic energy, thereby incorporating long-range Coulomb interactions into energy and force prediction. On three benchmark datasets, QM9, revised MD17, and tmQM_wB97MV, CE-ETNet reduces the mean absolute error (MAE) of energy and force prediction by about 8% on average compared with existing models, reaching chemical accuracy. The experimental results show that explicit long-range electrostatic modeling improves the predictive performance of equivariant graph neural network interatomic potentials and provides a useful design direction for machine-learning interatomic potential models.
References
- [1] Unke O T, Chmiela S, Sauceda H E, et al. Machine Learning Force Fields[J]. Chemical Reviews, 2021, 121(16): 10142-10186.
- [2] Anstine D M, Isayev O. Machine Learning Interatomic Potentials and Long-Range Physics[J]. The Journal of Physical Chemistry A, 2023, 127(11): 2417-2431.
- [3] Schütt K T, Kindermans P J, Sauceda H E, et al. SchNet: A continuous-filter convolutional neural network for modeling quantum interactions[C]. Advances in Neural Information Processing Systems, 2017: 991-1001.
- [4] Gasteiger J, Groß J, Günnemann S. Directional Message Passing for Molecular Graphs[C]. International Conference on Learning Representations, 2020.
- [5] Batzner S, Musaelian A, Sun L, et al. E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials[J]. Nature Communications, 2022, 13: 2453.
- [6] Schütt K T, Unke O T, Gastegger M. Equivariant message passing for the prediction of tensorial properties and molecular spectra[C]. International Conference on Machine Learning, 2021: 9377-9388.
- [7] Rappé A K, Goddard W A. Charge equilibration for molecular dynamics simulations[J]. The Journal of Physical Chemistry, 1991, 95(8): 3358-3363.
- [8] Unke O T, Meuwly M. PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges[J]. Journal of Chemical Theory and Computation, 2019, 15(6): 3678-3693.
- [9] Unke O T, Chmiela S, Gastegger M, et al. SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects[J]. Nature Communications, 2021, 12: 7273.
- [10] Musaelian A, Batzner S, Johansson A, et al. Learning local equivariant representations for large-scale atomistic dynamics[J]. Nature Communications, 2023, 14: 579.
- [11] Thölke P, De Fabritiis G. Equivariant Transformers for Neural Network Based Molecular Potentials[C]. International Conference on Learning Representations, 2022.
- [12] Liao Y L, Smidt T. Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs[C]. International Conference on Learning Representations, 2023.
- [13] Batatia I, Kovács D P, Simm G, et al. MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields[C]. Advances in Neural Information Processing Systems, 2022.
- [14] Aykent S, Xia T. GotenNet: Rethinking Efficient 3D Equivariant Graph Neural Networks[C]. International Conference on Learning Representations, 2025.
- [15] Fuchs P, Sanocki M, Zavadlav J. Learning non-local molecular interactions via equivariant local representations and charge equilibration[J]. npj Computational Materials, 2025, 11: 287.
- [16] Vondrák M, Baldwin W J, Csányi G, et al. Integrating Charge Equilibration with Equivariant Machine-Learning Interatomic Potentials[J]. Journal of Chemical Theory and Computation, 2026, 22(13): 6874-6889.
- [17] Maruf M U, Kim S, Ahmad Z. Learning Long-Range Interactions in Equivariant Machine Learning Interatomic Potentials via Electronic Degrees of Freedom[J]. The Journal of Physical Chemistry Letters, 2025. DOI: 10.1021/acs.jpclett.5c02352.
- [18] Ramakrishnan R, Dral P O, Rupp M, von Lilienfeld O A. Quantum chemistry structures and properties of 134 kilo molecules[J]. Scientific Data, 2014, 1: 140022.
- [19] Christensen A S, von Lilienfeld O A. Revised MD17 dataset[DB/OL]. Materials Cloud Archive, 2020. DOI: 10.24435/materialscloud:wy-kn.
- [20] Garrison A G, Heras-Domingo J, Kitchin J R, et al. Applying Large Graph Neural Networks to Predict Transition Metal Complex Energies Using the tmQM_wB97MV Data Set[J]. Journal of Chemical Information and Modeling, 2023, 63: 7642-7654.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.