Abstract:High-throughput computational screening has emerged as a critical component of materials discovery. Direct density functional theory (DFT) simulation of inorganic materials and molecular transition metal complexes is often used to describe subtle trends in inorganic bonding and spin-state ordering, but these calculations are computationally costly and properties are sensitive to the exchange-correlation functional employed. To begin to overcome these challenges, we trained artificial neural networks (ANNs) to predict quantum-mechanically-derived properties, including spin-state ordering, sensitivity to Hartree-Fock exchange, and spin- state specific bond lengths in transition metal complexes. Our ANN is trained on a small set of inorganic-chemistry-appropriate empirical inputs that are both maximally transferable and do not require precise three-dimensional structural information for prediction. Using these descriptors, our ANN predicts spin-state splittings of single-site transition metal complexes (i.e., Cr-Ni) at arbitrary amounts of Hartree-Fock exchange to within 3 kcal/mol accuracy of DFT calculations. Our exchange-sensitivity ANN enables improved predictions on a diverse test set of experimentally-characterized transition metal complexes by extrapolation from semi-local DFT to hybrid DFT. The ANN also outperforms other machine learning models (i.e., support vector regression and kernel ridge regression), demonstrating particularly improved performance in transferability, as measured by prediction errors on the diverse test set. We establish the value of new uncertainty quantification tools to estimate ANN prediction uncertainty in computational chemistry, and we provide additional heuristics for identification of when a compound of interest is likely to be poorly predicted by the ANN.

Accurate prediction of grain boundary structures and energetics in CdTe: a machine-learning potential approach

Artificial neural network molecular mechanics of iron grain boundaries

Grain Boundaries Induce Significant Decrease in Lattice Thermal Conductivity of CdTe

Quantitative prediction of grain boundary thermal conductivities from local atomic environments

Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks

Unraveling the Hall-Petch to Inverse Hall-Petch Transition in Nanocrystalline CdTe

Machine learning interatomic potential with DFT accuracy for general grain boundaries in α-Fe

Exploring Large-Lattice-Mismatched Interfaces with Neural Network Potentials: the Case of the CdS/CdTe Heterostructure

Predicting Grain Boundary Structure and Energy in BCC Metals by Integrated Atomistic and Phase-Field Modeling

Standard Deviation Effect of Average Structure Descriptor on Grain Boundary Energy Prediction

Defect levels in d-electron containing systems: Comparative study of CdTe using LDA and LDA

Transferable Machine Learning Interatomic Potential for Carbon Hydrogen Systems

Learning from metastable grain boundaries

Prediction of vacancy formation energies at tungsten grain boundaries from local structure via machine learning method

A machine learning potential-based generative algorithm for on-lattice crystal structure prediction

Machine learning determination of atomic dynamics at grain boundaries

Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations

Directly Determining the Interface Structure and Band Offset of a Large-Lattice-Mismatched CdS/CdTe Heterostructure

Predicting grain boundary damage by machine learning

Constrained Bayesian Nonparametric Regression for Grain Boundary Energy Predictions

Fundamental Microscopic Properties as Predictors of Large-Scale Quantities of Interest: Validation through Grain Boundary Energy Trends