Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning
Abstract The chemical and structural properties of atomically precise nanoclusters are of great interest in numerous applications, but predicting the stable structures of clusters can be computationally expensive. In this work, we present a procedure for rapidly predicting low-energy structures of n...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-08-01
|
| Series: | npj Computational Materials |
| Online Access: | https://doi.org/10.1038/s41524-022-00856-x |
| _version_ | 1811340276403273728 |
|---|---|
| author | Yunzhe Wang Shanping Liu Peter Lile Sam Norwood Alberto Hernandez Sukriti Manna Tim Mueller |
| author_facet | Yunzhe Wang Shanping Liu Peter Lile Sam Norwood Alberto Hernandez Sukriti Manna Tim Mueller |
| author_sort | Yunzhe Wang |
| collection | DOAJ |
| description | Abstract The chemical and structural properties of atomically precise nanoclusters are of great interest in numerous applications, but predicting the stable structures of clusters can be computationally expensive. In this work, we present a procedure for rapidly predicting low-energy structures of nanoclusters by combining a genetic algorithm with interatomic potentials actively learned on-the-fly. Applying this approach to aluminum clusters with 21 to 55 atoms, we have identified structures with lower energy than any reported in the literature for 25 out of the 35 sizes. Our benchmarks indicate that the active learning procedure accelerated the average search speed by about an order of magnitude relative to genetic algorithm searches using only density functional calculations. This work demonstrates a feasible way to systematically discover stable structures for large nanoclusters and provides insights into the transferability of machine-learned interatomic potentials for nanoclusters. |
| first_indexed | 2024-04-13T18:39:29Z |
| format | Article |
| id | doaj.art-0f81a2c9e5204869abd15c212f9283ab |
| institution | Directory Open Access Journal |
| issn | 2057-3960 |
| language | English |
| last_indexed | 2024-04-13T18:39:29Z |
| publishDate | 2022-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Computational Materials |
| spelling | doaj.art-0f81a2c9e5204869abd15c212f9283ab2022-12-22T02:34:46ZengNature Portfolionpj Computational Materials2057-39602022-08-018111010.1038/s41524-022-00856-xAccelerated prediction of atomically precise cluster structures using on-the-fly machine learningYunzhe Wang0Shanping Liu1Peter Lile2Sam Norwood3Alberto Hernandez4Sukriti Manna5Tim Mueller6Department of Materials Science and Engineering, Johns Hopkins UniversityDepartment of Materials Science and Engineering, Johns Hopkins UniversityDepartment of Materials Science and Engineering, Johns Hopkins UniversityDepartment of Materials Science and Engineering, Johns Hopkins UniversityDepartment of Materials Science and Engineering, Johns Hopkins UniversityDepartment of Materials Science and Engineering, Johns Hopkins UniversityDepartment of Materials Science and Engineering, Johns Hopkins UniversityAbstract The chemical and structural properties of atomically precise nanoclusters are of great interest in numerous applications, but predicting the stable structures of clusters can be computationally expensive. In this work, we present a procedure for rapidly predicting low-energy structures of nanoclusters by combining a genetic algorithm with interatomic potentials actively learned on-the-fly. Applying this approach to aluminum clusters with 21 to 55 atoms, we have identified structures with lower energy than any reported in the literature for 25 out of the 35 sizes. Our benchmarks indicate that the active learning procedure accelerated the average search speed by about an order of magnitude relative to genetic algorithm searches using only density functional calculations. This work demonstrates a feasible way to systematically discover stable structures for large nanoclusters and provides insights into the transferability of machine-learned interatomic potentials for nanoclusters.https://doi.org/10.1038/s41524-022-00856-x |
| spellingShingle | Yunzhe Wang Shanping Liu Peter Lile Sam Norwood Alberto Hernandez Sukriti Manna Tim Mueller Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning npj Computational Materials |
| title | Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning |
| title_full | Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning |
| title_fullStr | Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning |
| title_full_unstemmed | Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning |
| title_short | Accelerated prediction of atomically precise cluster structures using on-the-fly machine learning |
| title_sort | accelerated prediction of atomically precise cluster structures using on the fly machine learning |
| url | https://doi.org/10.1038/s41524-022-00856-x |
| work_keys_str_mv | AT yunzhewang acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning AT shanpingliu acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning AT peterlile acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning AT samnorwood acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning AT albertohernandez acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning AT sukritimanna acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning AT timmueller acceleratedpredictionofatomicallypreciseclusterstructuresusingontheflymachinelearning |