Machine Learning Prediction of Charged Defect Formation Energies from Crystal Structures

Abstract

Recent advances in materials informatics have expanded the number of synthesizable materials. However, screening promising candidates, such as semiconductors, based on defect properties remains challenging. This is primarily due to the lack of a general framework for predicting defect formation energies in multiple charge states from structural data. In this Letter, we present a protocol, namely data normalization, Fermi level alignment, and treatment of perturbed host states, and validate it by accurately predicting oxygen vacancy formation energies in three charge states using a single model. We also introduce a joint machine-learning model that integrates defect formation energies and band-edge predictions for virtual screening. Using this framework, we identify 89 hole-dopable oxides, including BaGaSbO, a potential ambipolar photovoltaic material. Our protocol is expected to become a standard approach for machine-learning studies on point defect formation energies.

Published in Phys. Rev. Lett

References in corpus (7)

• Restoring the density-gradient expansion for exchange in solids and surfaces
• Generalized gradient approximation for solids and their surfaces
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Anion vacancies as a source of persistent photoconductivity in II-VI and chalcopyrite semiconductors
• Self-consistent hybrid functional for condensed systems
• Electrically Benign Defect Behavior in Zinc Tin Nitride Revealed from First Principles
• Investigation of Hole Dopability in Oxygen $2p$-Dominated Bands

Cited by in corpus (2)

• Accelerating point defect simulations using data-driven and machine learning approaches
• Oxygen-deficiency-driven phase segregation enables enhanced hole transport in amorphous tellurium oxides