CHGNet: Pretrained universal neural network potential for charge-informed atomistic modeling

TL;DR

Large-scale simulations of materials with complex electron interactions are limited by the $O(N^3)$ scaling of DFT and the cost of ab initio MD. CHGNet is a pretrained universal neural network potential that incorporates electronic degrees of freedom by using atomic magnetic moments as charge proxies, learned from the Materials Project Trajectory Dataset to predict energies, forces, stresses, and magmoms. It enables charge-informed MD and thermodynamic studies across $Li_{x}MnO_{2}$, $Li_{x}FePO_{4}$, and garnet conductors, revealing charge-disproportionation dynamics, electronic entropy effects, and activated diffusion networks beyond the reach of traditional MLIPs. By integrating electronic information into a scalable, graph-based framework, CHGNet offers a data-efficient platform for studying heterovalent systems and charge-transfer phenomena in materials science.

Abstract

The simulation of large-scale systems with complex electron interactions remains one of the greatest challenges for the atomistic modeling of materials. Although classical force fields often fail to describe the coupling between electronic states and ionic rearrangements, the more accurate \textit{ab-initio} molecular dynamics suffers from computational complexity that prevents long-time and large-scale simulations, which are essential to study many technologically relevant phenomena, such as reactions, ion migrations, phase transformations, and degradation. In this work, we present the Crystal Hamiltonian Graph neural Network (CHGNet) as a novel machine-learning interatomic potential (MLIP), using a graph-neural-network-based force field to model a universal potential energy surface. CHGNet is pretrained on the energies, forces, stresses, and magnetic moments from the Materials Project Trajectory Dataset, which consists of over 10 years of density functional theory static and relaxation trajectories of $\sim 1.5$ million inorganic structures. The explicit inclusion of magnetic moments enables CHGNet to learn and accurately represent the orbital occupancy of electrons, enhancing its capability to describe both atomic and electronic degrees of freedom. We demonstrate several applications of CHGNet in solid-state materials, including charge-informed molecular dynamics in Li$_x$MnO$_2$, the finite temperature phase diagram for Li$_x$FePO$_4$ and Li diffusion in garnet conductors. We critically analyze the significance of including charge information for capturing appropriate chemistry, and we provide new insights into ionic systems with additional electronic degrees of freedom that can not be observed by previous MLIPs.

Figures (6)

• Figure 1: CHGNet model architecture (a) CHGNet workflow: a crystal structure with unknown atomic charge is used as input to predict the energy, force, stress, and magnetic moments, resulting in a charge-decorated structure. (b) Atom graph: The pairwise bond information is drawn between atoms; Bond graph: the pairwise angle information is drawn between bonds. (c) Graphs run through basis expansions and embedding layers to create atom, bond, angle features. The features are updated through several interaction blocks, and the properties are predicted at output layers. (d) Interaction block in which the atom, bond, and angle share and update information. (e) Atom convolution layer where neighboring atom and bond information is calculated through weighted message passing and aggregates to the atoms.
• Figure 2: Element distribution of Materials Project Trajectory (MPtrj) Dataset. The color on the lower-left triangle indicates the total number of atoms/ions of an element. The color on the upper right indicates the number of times the atoms/ions are incorporated with magnetic moment labels in the MPtrj dataset. On the lower part of the plot is the count and mean absolute deviation (MAD) of energy, magmoms, force, and stress
• Figure 3: Magmom and hidden-space regularization in Na$_2$V$_2$(PO$_4$)$_3$. (a) Magmom distribution of the 216 V ions in the unrelaxed structure (blue) and CHGNet-relaxed structure (orange). (b) A two-dimensional visualization of the PCA on V-ion embedding vectors before the magmom projection layer indicates the latent space clustering is highly correlated with magmoms and charge information. The PCA reduction is calculated for both unrelaxed and relaxed structures.
• Figure 4: Li$_{0.5}$MnO$_2$ phase transformation and charge disproportionation (a) orthorhombic LiMnO$_2$ (o-LMO) unit cell plotted with the tetrahedral site and the octahedral site. (b) Simulated XRD pattern of CHGNet MD structures as the system transforms from the o-LMO phase to the s-LMO. (c) Average magmoms of tetrahedral and octahedral Mn ions vs. time. (d) Top: total potential energy and the relative intensity of o-LMO and s-LMO characteristic peaks vs. time. The solid black line is averaged over every 10 ps. Bottom: the histogram of magmoms on all Mn ions vs. time. The brighter color indicates more Mn ions distributed at the magmom. (e) Predicted magmoms of tetrahedral Mn ions using GGA$+U$-DFT (black) and CHGNet (blue), where the structures are drawn from MD simulation at 0.4 ns (left) and 1.5 ns (right).
• Figure 5: Li$_x$FePO$_4$ phase diagram from CHGNet. The phase diagrams in (a) and (b) are calculated with and without electronic entropy on Fe$^{2+}$ and Fe$^{3+}$. The colored dots represent the stable phases obtained in semi-grand canonical MC. The dashed lines indicate the two-phase equilibria between solid solution phases.