Energy Materials Screening with Defect Graph Neural Networks

Matthew Witman, Stephan Lany

Research output: Contribution to journalArticle


Graph neural networks (GNNs) present a promising route for machine learning of solid-state materials' properties, but methods capable of directly predicting defect properties from ideal, defect-free structures are needed. A GNN developed for direct defect property predictions enables high-throughput screening of redox-active oxides for energy applications and beyond.
Original languageAmerican English
Pages (from-to)671-672
Number of pages2
JournalNature Computational Science
StatePublished - 2023

Bibliographical note

See NREL/JA-5K00-84167 for related paper

NREL Publication Number

  • NREL/JA-5K00-86991


  • machine learning
  • solar thermochemical hydrogen


Dive into the research topics of 'Energy Materials Screening with Defect Graph Neural Networks'. Together they form a unique fingerprint.

Cite this