Probabilistic phase labeling and lattice refinement for autonomous materials research
Abstract X-ray diffraction (XRD) is a powerful method for determining a material’s crystal structure in high-throughput experimentation, and is widely being incorporated in artificially intelligent agents for autonomous scientific discovery. However, rapid, automated, and reliable analysis of XRD da...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-05-01
|
| Series: | npj Computational Materials |
| Online Access: | https://doi.org/10.1038/s41524-025-01627-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849730962052612096 |
|---|---|
| author | Ming-Chiang Chang Sebastian Ament Maximilian Amsler Duncan R. Sutherland Lan Zhou John M. Gregoire Carla P. Gomes R. Bruce van Dover Michael O. Thompson |
| author_facet | Ming-Chiang Chang Sebastian Ament Maximilian Amsler Duncan R. Sutherland Lan Zhou John M. Gregoire Carla P. Gomes R. Bruce van Dover Michael O. Thompson |
| author_sort | Ming-Chiang Chang |
| collection | DOAJ |
| description | Abstract X-ray diffraction (XRD) is a powerful method for determining a material’s crystal structure in high-throughput experimentation, and is widely being incorporated in artificially intelligent agents for autonomous scientific discovery. However, rapid, automated, and reliable analysis of XRD data at rates that match the pace of experimental measurements at a synchrotron source remains a major challenge. To address these issues, we developed CrystalShift for rapid and efficient probabilistic XRD phase labeling employing symmetry-constrained optimization, best-first tree search, and Bayesian model comparison. The algorithm estimates probabilities for phase combinations without requiring additional phase space information or training. We demonstrate that CrystalShift provides robust probability estimates, outperforming existing methods on synthetic and experimental datasets, and can be readily integrated into high-throughput experimental workflows. In addition to efficient phase labeling, CrystalShift offers quantitative insights into materials’ structural parameters, which facilitate both expert evaluation and AI-based modeling of the phase space, ultimately accelerating materials identification and discovery. |
| format | Article |
| id | doaj-art-34bf25da69fa420f946b00bbdf664695 |
| institution | DOAJ |
| issn | 2057-3960 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Computational Materials |
| spelling | doaj-art-34bf25da69fa420f946b00bbdf6646952025-08-20T03:08:43ZengNature Portfolionpj Computational Materials2057-39602025-05-0111111310.1038/s41524-025-01627-0Probabilistic phase labeling and lattice refinement for autonomous materials researchMing-Chiang Chang0Sebastian Ament1Maximilian Amsler2Duncan R. Sutherland3Lan Zhou4John M. Gregoire5Carla P. Gomes6R. Bruce van Dover7Michael O. Thompson8Department of Materials Science and Engineering, Cornell UniversityDepartment of Computer Science, Cornell UniversityDepartment of Materials Science and Engineering, Cornell UniversityDepartment of Materials Science and Engineering, Cornell UniversityDivision of Engineering and Applied Science, California Institute of TechnologyDivision of Engineering and Applied Science, California Institute of TechnologyDepartment of Computer Science, Cornell UniversityDepartment of Materials Science and Engineering, Cornell UniversityDepartment of Materials Science and Engineering, Cornell UniversityAbstract X-ray diffraction (XRD) is a powerful method for determining a material’s crystal structure in high-throughput experimentation, and is widely being incorporated in artificially intelligent agents for autonomous scientific discovery. However, rapid, automated, and reliable analysis of XRD data at rates that match the pace of experimental measurements at a synchrotron source remains a major challenge. To address these issues, we developed CrystalShift for rapid and efficient probabilistic XRD phase labeling employing symmetry-constrained optimization, best-first tree search, and Bayesian model comparison. The algorithm estimates probabilities for phase combinations without requiring additional phase space information or training. We demonstrate that CrystalShift provides robust probability estimates, outperforming existing methods on synthetic and experimental datasets, and can be readily integrated into high-throughput experimental workflows. In addition to efficient phase labeling, CrystalShift offers quantitative insights into materials’ structural parameters, which facilitate both expert evaluation and AI-based modeling of the phase space, ultimately accelerating materials identification and discovery.https://doi.org/10.1038/s41524-025-01627-0 |
| spellingShingle | Ming-Chiang Chang Sebastian Ament Maximilian Amsler Duncan R. Sutherland Lan Zhou John M. Gregoire Carla P. Gomes R. Bruce van Dover Michael O. Thompson Probabilistic phase labeling and lattice refinement for autonomous materials research npj Computational Materials |
| title | Probabilistic phase labeling and lattice refinement for autonomous materials research |
| title_full | Probabilistic phase labeling and lattice refinement for autonomous materials research |
| title_fullStr | Probabilistic phase labeling and lattice refinement for autonomous materials research |
| title_full_unstemmed | Probabilistic phase labeling and lattice refinement for autonomous materials research |
| title_short | Probabilistic phase labeling and lattice refinement for autonomous materials research |
| title_sort | probabilistic phase labeling and lattice refinement for autonomous materials research |
| url | https://doi.org/10.1038/s41524-025-01627-0 |
| work_keys_str_mv | AT mingchiangchang probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT sebastianament probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT maximilianamsler probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT duncanrsutherland probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT lanzhou probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT johnmgregoire probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT carlapgomes probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT rbrucevandover probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch AT michaelothompson probabilisticphaselabelingandlatticerefinementforautonomousmaterialsresearch |