Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns
Abstract Two-dimensional (2D) materials have garnered significant attention due to their tunable electronic and optical properties and exceptional mechanical performance. Reconstructing 2D structures from diffraction patterns without prior assumptions or comprehensive knowledge is challenging, espec...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-05-01
|
| Series: | Communications Physics |
| Online Access: | https://doi.org/10.1038/s42005-025-02152-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849688340678311936 |
|---|---|
| author | Rong Fu Tianhao Su Musen Li Yue Wu Runhai Ouyang Danica Solina Michael Cortie Tongyi Zhang Shunbo Hu Zhongming Ren |
| author_facet | Rong Fu Tianhao Su Musen Li Yue Wu Runhai Ouyang Danica Solina Michael Cortie Tongyi Zhang Shunbo Hu Zhongming Ren |
| author_sort | Rong Fu |
| collection | DOAJ |
| description | Abstract Two-dimensional (2D) materials have garnered significant attention due to their tunable electronic and optical properties and exceptional mechanical performance. Reconstructing 2D structures from diffraction patterns without prior assumptions or comprehensive knowledge is challenging, especially for heterogeneous stacking and quantum 2D materials. Here, we introduce DD2D (diffraction pattern deep-reconstruction 2D structures), a physics-guided deep learning method that predicts 2D structures directly from diffraction patterns. DD2D employs a twin-tower framework, integrating a crystallographic geometric encoder and a site texture encoder, and uses a self-attention mechanism to identify intrinsic correlations in physical information and corresponding areas in the diffraction pattern. The results demonstrate high anti-interference, robust recognition capabilities, reliable interpretability, and prediction accuracy of up to 99.0%, highlighting its potential for future 2D materials discoveries. |
| format | Article |
| id | doaj-art-e556841b87754347aae031f1d73a3ff7 |
| institution | DOAJ |
| issn | 2399-3650 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Physics |
| spelling | doaj-art-e556841b87754347aae031f1d73a3ff72025-08-20T03:22:03ZengNature PortfolioCommunications Physics2399-36502025-05-01811710.1038/s42005-025-02152-8Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patternsRong Fu0Tianhao Su1Musen Li2Yue Wu3Runhai Ouyang4Danica Solina5Michael Cortie6Tongyi Zhang7Shunbo Hu8Zhongming Ren9State Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversitySchool of Mathematical and Physical Sciences, University of Technology Sydney, New South WalesSchool of Mathematical and Physical Sciences, University of Technology Sydney, New South WalesState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityState Key Laboratory of Advanced Special Steel, School of Materials Science and Engineering & Materials Genome Institute, Shanghai UniversityAbstract Two-dimensional (2D) materials have garnered significant attention due to their tunable electronic and optical properties and exceptional mechanical performance. Reconstructing 2D structures from diffraction patterns without prior assumptions or comprehensive knowledge is challenging, especially for heterogeneous stacking and quantum 2D materials. Here, we introduce DD2D (diffraction pattern deep-reconstruction 2D structures), a physics-guided deep learning method that predicts 2D structures directly from diffraction patterns. DD2D employs a twin-tower framework, integrating a crystallographic geometric encoder and a site texture encoder, and uses a self-attention mechanism to identify intrinsic correlations in physical information and corresponding areas in the diffraction pattern. The results demonstrate high anti-interference, robust recognition capabilities, reliable interpretability, and prediction accuracy of up to 99.0%, highlighting its potential for future 2D materials discoveries.https://doi.org/10.1038/s42005-025-02152-8 |
| spellingShingle | Rong Fu Tianhao Su Musen Li Yue Wu Runhai Ouyang Danica Solina Michael Cortie Tongyi Zhang Shunbo Hu Zhongming Ren Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns Communications Physics |
| title | Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns |
| title_full | Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns |
| title_fullStr | Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns |
| title_full_unstemmed | Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns |
| title_short | Physics-guided deep learning strategy for 2D structure reconstruction from diffraction patterns |
| title_sort | physics guided deep learning strategy for 2d structure reconstruction from diffraction patterns |
| url | https://doi.org/10.1038/s42005-025-02152-8 |
| work_keys_str_mv | AT rongfu physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT tianhaosu physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT musenli physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT yuewu physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT runhaiouyang physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT danicasolina physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT michaelcortie physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT tongyizhang physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT shunbohu physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns AT zhongmingren physicsguideddeeplearningstrategyfor2dstructurereconstructionfromdiffractionpatterns |