Cargando…
Machine learning magnetism classifiers from atomic coordinates
The determination of magnetic structure poses a long-standing challenge in condensed matter physics and materials science. Experimental techniques such as neutron diffraction are resource-limited and require complex structure refinement protocols, while computational approaches such as first-princip...
Autores principales: | , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Elsevier
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9574499/ https://www.ncbi.nlm.nih.gov/pubmed/36262309 http://dx.doi.org/10.1016/j.isci.2022.105192 |
_version_ | 1784811116567724032 |
---|---|
author | Merker, Helena A. Heiberger, Harry Nguyen, Linh Liu, Tongtong Chen, Zhantao Andrejevic, Nina Drucker, Nathan C. Okabe, Ryotaro Kim, Song Eun Wang, Yao Smidt, Tess Li, Mingda |
author_facet | Merker, Helena A. Heiberger, Harry Nguyen, Linh Liu, Tongtong Chen, Zhantao Andrejevic, Nina Drucker, Nathan C. Okabe, Ryotaro Kim, Song Eun Wang, Yao Smidt, Tess Li, Mingda |
author_sort | Merker, Helena A. |
collection | PubMed |
description | The determination of magnetic structure poses a long-standing challenge in condensed matter physics and materials science. Experimental techniques such as neutron diffraction are resource-limited and require complex structure refinement protocols, while computational approaches such as first-principles density functional theory (DFT) need additional semi-empirical correction, and reliable prediction is still largely limited to collinear magnetism. Here, we present a machine learning model that aims to classify the magnetic structure by inputting atomic coordinates containing transition metal and rare earth elements. By building a Euclidean equivariant neural network that preserves the crystallographic symmetry, the magnetic structure (ferromagnetic, antiferromagnetic, and non-magnetic) and magnetic propagation vector (zero or non-zero) can be predicted with an average accuracy of 77.8% and 73.6%. In particular, a 91% accuracy is reached when predicting no magnetic ordering even if the structure contains magnetic element(s). Our work represents one step forward to solving the grand challenge of full magnetic structure determination. |
format | Online Article Text |
id | pubmed-9574499 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | Elsevier |
record_format | MEDLINE/PubMed |
spelling | pubmed-95744992022-10-18 Machine learning magnetism classifiers from atomic coordinates Merker, Helena A. Heiberger, Harry Nguyen, Linh Liu, Tongtong Chen, Zhantao Andrejevic, Nina Drucker, Nathan C. Okabe, Ryotaro Kim, Song Eun Wang, Yao Smidt, Tess Li, Mingda iScience Article The determination of magnetic structure poses a long-standing challenge in condensed matter physics and materials science. Experimental techniques such as neutron diffraction are resource-limited and require complex structure refinement protocols, while computational approaches such as first-principles density functional theory (DFT) need additional semi-empirical correction, and reliable prediction is still largely limited to collinear magnetism. Here, we present a machine learning model that aims to classify the magnetic structure by inputting atomic coordinates containing transition metal and rare earth elements. By building a Euclidean equivariant neural network that preserves the crystallographic symmetry, the magnetic structure (ferromagnetic, antiferromagnetic, and non-magnetic) and magnetic propagation vector (zero or non-zero) can be predicted with an average accuracy of 77.8% and 73.6%. In particular, a 91% accuracy is reached when predicting no magnetic ordering even if the structure contains magnetic element(s). Our work represents one step forward to solving the grand challenge of full magnetic structure determination. Elsevier 2022-09-28 /pmc/articles/PMC9574499/ /pubmed/36262309 http://dx.doi.org/10.1016/j.isci.2022.105192 Text en © 2022 The Author(s) https://creativecommons.org/licenses/by-nc-nd/4.0/This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). |
spellingShingle | Article Merker, Helena A. Heiberger, Harry Nguyen, Linh Liu, Tongtong Chen, Zhantao Andrejevic, Nina Drucker, Nathan C. Okabe, Ryotaro Kim, Song Eun Wang, Yao Smidt, Tess Li, Mingda Machine learning magnetism classifiers from atomic coordinates |
title | Machine learning magnetism classifiers from atomic coordinates |
title_full | Machine learning magnetism classifiers from atomic coordinates |
title_fullStr | Machine learning magnetism classifiers from atomic coordinates |
title_full_unstemmed | Machine learning magnetism classifiers from atomic coordinates |
title_short | Machine learning magnetism classifiers from atomic coordinates |
title_sort | machine learning magnetism classifiers from atomic coordinates |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9574499/ https://www.ncbi.nlm.nih.gov/pubmed/36262309 http://dx.doi.org/10.1016/j.isci.2022.105192 |
work_keys_str_mv | AT merkerhelenaa machinelearningmagnetismclassifiersfromatomiccoordinates AT heibergerharry machinelearningmagnetismclassifiersfromatomiccoordinates AT nguyenlinh machinelearningmagnetismclassifiersfromatomiccoordinates AT liutongtong machinelearningmagnetismclassifiersfromatomiccoordinates AT chenzhantao machinelearningmagnetismclassifiersfromatomiccoordinates AT andrejevicnina machinelearningmagnetismclassifiersfromatomiccoordinates AT druckernathanc machinelearningmagnetismclassifiersfromatomiccoordinates AT okaberyotaro machinelearningmagnetismclassifiersfromatomiccoordinates AT kimsongeun machinelearningmagnetismclassifiersfromatomiccoordinates AT wangyao machinelearningmagnetismclassifiersfromatomiccoordinates AT smidttess machinelearningmagnetismclassifiersfromatomiccoordinates AT limingda machinelearningmagnetismclassifiersfromatomiccoordinates |