Dielectric tensor prediction for inorganic materials using latent information from preferred potential

Abstract Dielectrics are crucial for technologies like flash memory, CPUs, photovoltaics, and capacitors, but public data on these materials are scarce, restricting research and development. Existing machine learning models have focused on predicting scalar polycrystalline dielectric constants, negl...

Full description

Saved in:
Bibliographic Details
Main Authors: Zetian Mao, WenWen Li, Jethro Tan
Format: Article
Language:English
Published: Nature Portfolio 2024-11-01
Series:npj Computational Materials
Online Access:https://doi.org/10.1038/s41524-024-01450-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850163299837018112
author Zetian Mao
WenWen Li
Jethro Tan
author_facet Zetian Mao
WenWen Li
Jethro Tan
author_sort Zetian Mao
collection DOAJ
description Abstract Dielectrics are crucial for technologies like flash memory, CPUs, photovoltaics, and capacitors, but public data on these materials are scarce, restricting research and development. Existing machine learning models have focused on predicting scalar polycrystalline dielectric constants, neglecting the directional nature of dielectric tensors essential for material design. This study leverages multi-rank equivariant structural embeddings from a universal neural network potential to enhance predictions of dielectric tensors. We develop an equivariant readout decoder to predict total, electronic, and ionic dielectric tensors while preserving O(3) equivariance, and benchmark its performance against state-of-the-art algorithms. Virtual screening of thermodynamically stable materials from Materials Project for two discovery tasks, high-dielectric and highly anisotropic materials, identifies promising candidates including Cs2Ti(WO4)3 (band gap E g = 2.93eV, dielectric constant ε = 180.90) and CsZrCuSe3 (anisotropic ratio α r = 121.89). The results demonstrate our model’s accuracy in predicting dielectric tensors and its potential for discovering novel dielectric materials.
format Article
id doaj-art-b568515d3da74d508df2ab4708dfdce8
institution OA Journals
issn 2057-3960
language English
publishDate 2024-11-01
publisher Nature Portfolio
record_format Article
series npj Computational Materials
spelling doaj-art-b568515d3da74d508df2ab4708dfdce82025-08-20T02:22:20ZengNature Portfolionpj Computational Materials2057-39602024-11-0110111510.1038/s41524-024-01450-zDielectric tensor prediction for inorganic materials using latent information from preferred potentialZetian Mao0WenWen Li1Jethro Tan2Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5 KashiwanohaPreferred Networks Inc., 1-6-1 Otemachi, Chiyoda-kuPreferred Networks Inc., 1-6-1 Otemachi, Chiyoda-kuAbstract Dielectrics are crucial for technologies like flash memory, CPUs, photovoltaics, and capacitors, but public data on these materials are scarce, restricting research and development. Existing machine learning models have focused on predicting scalar polycrystalline dielectric constants, neglecting the directional nature of dielectric tensors essential for material design. This study leverages multi-rank equivariant structural embeddings from a universal neural network potential to enhance predictions of dielectric tensors. We develop an equivariant readout decoder to predict total, electronic, and ionic dielectric tensors while preserving O(3) equivariance, and benchmark its performance against state-of-the-art algorithms. Virtual screening of thermodynamically stable materials from Materials Project for two discovery tasks, high-dielectric and highly anisotropic materials, identifies promising candidates including Cs2Ti(WO4)3 (band gap E g = 2.93eV, dielectric constant ε = 180.90) and CsZrCuSe3 (anisotropic ratio α r = 121.89). The results demonstrate our model’s accuracy in predicting dielectric tensors and its potential for discovering novel dielectric materials.https://doi.org/10.1038/s41524-024-01450-z
spellingShingle Zetian Mao
WenWen Li
Jethro Tan
Dielectric tensor prediction for inorganic materials using latent information from preferred potential
npj Computational Materials
title Dielectric tensor prediction for inorganic materials using latent information from preferred potential
title_full Dielectric tensor prediction for inorganic materials using latent information from preferred potential
title_fullStr Dielectric tensor prediction for inorganic materials using latent information from preferred potential
title_full_unstemmed Dielectric tensor prediction for inorganic materials using latent information from preferred potential
title_short Dielectric tensor prediction for inorganic materials using latent information from preferred potential
title_sort dielectric tensor prediction for inorganic materials using latent information from preferred potential
url https://doi.org/10.1038/s41524-024-01450-z
work_keys_str_mv AT zetianmao dielectrictensorpredictionforinorganicmaterialsusinglatentinformationfrompreferredpotential
AT wenwenli dielectrictensorpredictionforinorganicmaterialsusinglatentinformationfrompreferredpotential
AT jethrotan dielectrictensorpredictionforinorganicmaterialsusinglatentinformationfrompreferredpotential