Energy Materials Screening with Defect Graph Neural Networks

Matthew Witman; Stephan Lany

doi:10.1038/s43588-023-00510-6

Energy Materials Screening with Defect Graph Neural Networks

Matthew Witman, Stephan Lany

Materials Science

Sandia National Laboratories

Research output: Contribution to journal › Article

Abstract

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 language	American English
Pages (from-to)	671-672
Number of pages	2
Journal	Nature Computational Science
Volume	3
DOIs	https://doi.org/10.1038/s43588-023-00510-6
State	Published - 2023

Bibliographical note

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

NREL Publication Number

NREL/JA-5K00-86991

Keywords

machine learning
solar thermochemical hydrogen

Access to Document

10.1038/s43588-023-00510-6

Cite this

@article{cdc965da34524811817a886bbbc137c3,

title = "Energy Materials Screening with Defect Graph Neural Networks",

abstract = "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.",

keywords = "machine learning, solar thermochemical hydrogen",

author = "Matthew Witman and Stephan Lany",

note = "See NREL/JA-5K00-84167 for related paper",

year = "2023",

doi = "10.1038/s43588-023-00510-6",

language = "American English",

volume = "3",

pages = "671--672",

journal = "Nature Computational Science",

issn = "2662-8457",

publisher = "Springer Nature",

}

TY - JOUR

T1 - Energy Materials Screening with Defect Graph Neural Networks

AU - Witman, Matthew

AU - Lany, Stephan

N1 - See NREL/JA-5K00-84167 for related paper

PY - 2023

Y1 - 2023

N2 - 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.

AB - 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.

KW - machine learning

KW - solar thermochemical hydrogen

U2 - 10.1038/s43588-023-00510-6

DO - 10.1038/s43588-023-00510-6

M3 - Article

SN - 2662-8457

VL - 3

SP - 671

EP - 672

JO - Nature Computational Science

JF - Nature Computational Science

ER -

Energy Materials Screening with Defect Graph Neural Networks

Abstract

Bibliographical note

NREL Publication Number

Keywords

Access to Document

Fingerprint

Cite this