A Unified Graph Neural Network Framework for Non-Equilibrium Carrier and Lattice Dynamics Driven by Electric Fields

2026/08/04 by Jia-Wen Li, Sheng Meng, Xinghua Shi +2
Physics and Astronomy · #physics.comp-ph

paper · pdf

arxiv created 2026/08/04 · arxiv updated 2026/08/05

Abstract

Finite-temperature simulations of electric-field-driven dynamics need a unified description of interatomic interactions, local electronic states, and configuration-dependent electric responses. First-principles simulations remain scale-limited, whereas conventional machine-learning potentials lack electric-field effects. Recent machine-learning frameworks have incorporated electric-field response or atom-resolved electronic-state information, but rarely both within a single framework. Here, we develop an electric-field-response graph neural network (EFR-GNN) that predicts energies, forces, Born effective charge tensors, atom-resolved charges and magnetic moments, and supports long-time field-driven molecular dynamics with atom-resolved tracking of localized electronic states. In hole-doped MgO, static fields rectify thermally activated hole-polaron hopping through a forward--backward asymmetry quantified by a nearest-neighbor model. In GaAs, resonant terahertz excitation generates a coherent Γ-point transverse-optical phonon with dephasing consistent with experiment, while opposite helicities reverse its rotation. In superionic α-AgI, it reproduces temperature-dependent Ag⁺ mobility and collective field-driven ionic drift. Together, EFR-GNN offers an approach to finite-temperature simulations of field-driven atomic and localized-carrier dynamics.

Citations

• Accurate Prediction of Tensorial Spectra Using Equivariant Graph Neural Network
• A Universal Spin-Orbit-Coupled Hamiltonian Model for Accelerated Quantum Material Discovery
• Machine learning interatomic potential can infer electrical response
• Observation of angular momentum transfer among crystal lattice modes
• Machine Learning Small Polaron Dynamics
• DFT dataset for "Symmetry Breaking in the Superionic Phase of Silver-Iodide"
• Representing Born effective charges with equivariant graph convolutional neural networks
• Scalable Crystal Structure Relaxation Using an Iteration-Free Deep Generative Model with Uncertainty Quantification
• Unified Differentiable Learning of Electric Response
• Phononic Switching of Magnetization by the Ultrafast Barnett Effect
• Direct calculation of the ionic mobility in superionic conductors
• MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields
• The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentials
• Learning Local Equivariant Representations for Large-Scale Atomistic Dynamics
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Machine Learning for Molecular Simulation
• On-the-Fly Active Learning of Interpretable Bayesian Force Fields for Atomistic Rare Events
• The atomic simulation environment—a Python library for working with atoms
• An effective magnetic field from optically driven phonons
• Angular Momentum of Phonons and the Einstein–de Haas Effect
• Progress in Time-Dependent Density-Functional Theory
• Dynamical Structure, Bonding, and Thermodynamics of the Superionic Sublattice in alpha-AgI
• Electronic excitations: density-functional versus many-body Green’s-function approaches
• Phonons and related crystal properties from density-functional perturbation theory
• Electron-energy-loss spectra and the structural stability of nickel oxide: An LSDA+U study
• Generalized Gradient Approximation Made Simple
• Efficient iterative schemes forab initiototal-energy calculations using a plane-wave basis set
• Projector augmented-wave method