Abstract:Having access to accurate electron densities in chemical systems, especially for dynamical systems involving chemical reactions, ion transport, and other charge transfer processes, is crucial for numerous applications in materials chemistry. Traditional methods for computationally predicting electron density data for such systems include quantum mechanical (QM) techniques, such as density functional theory. However, poor scaling of these QM methods restricts their use to relatively small system sizes and short dynamic time scales. To overcome this limitation, we have developed a deep neural network machine learning formalism, which we call deep charge density prediction (DeepCDP), for predicting charge densities by only using atomic positions for molecules and condensed phase (periodic) systems. Our method uses the weighted smooth overlap of atomic positions to fingerprint environments on a grid-point basis and map it to electron density data generated from QM simulations. We trained models for bulk systems of copper, LiF, and silicon; for a molecular system, water; and for two-dimensional charged and uncharged systems, hydroxyl-functionalized graphane, with and without an added proton. We showed that DeepCDP achieves prediction R2 values greater than 0.99 and mean squared error values on the order of 10−5e2 Å−6 for most systems. DeepCDP scales linearly with system size, is highly parallelizable, and is capable of accurately predicting the excess charge in protonated hydroxyl-functionalized graphane. We demonstrate how DeepCDP can be used to accurately track the location of charges (protons) by computing electron densities at a few selected grid points in the materials, thus significantly reducing the computational cost. We also show that our models can be transferable, allowing prediction of electron densities for systems on which it has not been trained but that contain a subset of atomic species on which it has been trained. Our approach can be used to develop models that span different chemical systems and train them for the study of large-scale charge transport and chemical reactions.

Graph Nets for Partial Charge Prediction

Out-of-the-box deep learning prediction of quantum-mechanical partial charges by graph representation and transfer learning

DeepAtomicCharge: a new graph convolutional network-based architecture for accurate prediction of atomic charges

A Recipe for Charge Density Prediction

Charge-Optimized Electrostatic Interaction Atom-Centered Neural Network Algorithm

DASH: Dynamic Attention-Based Substructure Hierarchy for Partial Charge Assignment

DDEC6: A Method for Computing Even-Tempered Net Atomic Charges in Periodic and Nonperiodic Materials

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

Expanding the Scope of Density Derived Electrostatic and Chemical Charge Partitioning to Thousands of Atoms

CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling

Equivariant graph neural networks for fast electron density estimation of molecules, liquids, and solids

Learning Atomic Multipoles: Prediction of the Electrostatic Potential with Equivariant Graph Neural Networks

Higher-Order Equivariant Neural Networks for Charge Density Prediction in Materials

PACMAN: A Robust Partial Atomic Charge Predicter for Nanoporous Materials Based on Crystal Graph Convolution Networks

Fast and accurate prediction of partial charges using Atom-Path-Descriptor-based machine learning

Non-equilibrium molecular geometries in graph neural networks

Accurate Calculation of Interatomic Forces with Neural Networks Based on a Generative Transformer Architecture

Transferable and extensible machine learning derived atomic charges for modeling hybrid nanoporous materials

Machine Learning Electron Density Prediction Using Weighted Smooth Overlap of Atomic Positions

Chemical Properties from Graph Neural Network-Predicted Electron Densities

Graph-EAM: an Interpretable and Efficient Graph Neural Network Potential Framework.