Modeling crystal defects using defect-informed neural networks

Modeling crystal defects using defect-informed neural networks | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Modeling crystal defects using defect-informed neural networks Lei Shen, Ziduo Yang, Xiaoqing Liu, Xiuying Zhang, Pengru Huang, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6209794/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 15 Jul, 2025 Read the published version in npj Computational Materials → Version 1 posted 10 You are reading this latest preprint version Abstract Machine learning has revolutionized the study of crystalline materials forenabling rapid predictions and discovery. However, most AI-for-Materialsresearch to date has focused on ideal crystals, whereas real-world materialsinevitably contain defects that play a critical role in modern functionaltechnologies. The defects and dopants break geometric symmetry andincrease interaction complexity, posing particular challenges for traditionalML models. Addressing these challenges requires models that areable to capture sparse defect-driven effects in crystals while maintainingadaptability and precision. Here, we introduce Defect-Informed EquivariantGraph Neural Network (DefiNet), a model specifically designedto accurately capture defect-related interactions and geometric configurationsin point-defect structures. Trained on 14,866 defect structures,DefiNet achieves highly accurate structural predictions in a single step,avoiding the time-consuming iterative processes in modern ML relaxationmodels and possible error accumulation from iteration. We furthervalidates DefiNet’s accuracy by using density functional theory (DFT) relaxation on DefiNet-predicted structures. For most defect structures,regardless of defect complexity or system size, only 3 ionic steps arerequired to reach the DFT-level ground state. Finally, comparisonswith scanning transmission electron microscopy (STEM) images confirmDefiNet’s scalability and extrapolation beyond point defects, positioningit as a groundbreaking tool for defect-focused materials research. Physical sciences/Materials science/Theory and computation/Atomistic models Physical sciences/Physics/Condensed matter physics/Structure of solids and liquids Defect Calculations Materials Discovery Equivariant Graph Neural Networks Structural Relaxation Full Text Additional Declarations No competing interests reported. Supplementary Files SI.pdf Cite Share Download PDF Status: Published Journal Publication published 15 Jul, 2025 Read the published version in npj Computational Materials → Version 1 posted Editorial decision: Revision requested 22 Apr, 2025 Reviews received at journal 19 Apr, 2025 Reviews received at journal 18 Apr, 2025 Reviewers agreed at journal 08 Apr, 2025 Reviewers agreed at journal 02 Apr, 2025 Reviewers agreed at journal 28 Mar, 2025 Reviewers invited by journal 28 Mar, 2025 Editor assigned by journal 28 Mar, 2025 Submission checks completed at journal 25 Mar, 2025 First submitted to journal 12 Mar, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6209794","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":440313015,"identity":"732558cf-e219-47aa-b427-19a6b4fea383","order_by":0,"name":"Lei Shen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA30lEQVRIiWNgGAWjYDACCSB+wCAhB+GxEaslgUHCmGQtDIkNRGuRn9187EFChUX6don0Bwwfyg4zyPcvwK/F4M6xdIOEMxK5O2fkGDDOOHeYweDGAwJaJHLMJBLbJHI33MhhYOZtA2qROEDAYTPyv0kk/pNIN7iR/oD5L1CL/AwCWhhu5LBJJDZIJBjcSDBgZgRqYTjfQMBhN9LMJBKOSRhuOPPG4GDPuXQegxsELJGfkfxM4kNNnbzB8fSHD36UWcvJ9xNyGDIAqeVhkEggQQsE8JNiyygYBaNgFIwEAADJOUURpcXNRgAAAABJRU5ErkJggg==","orcid":"","institution":"National university of singapore","correspondingAuthor":true,"prefix":"","firstName":"Lei","middleName":"","lastName":"Shen","suffix":""},{"id":440313016,"identity":"f7d1e9ab-6373-42a8-9972-5b55408c415b","order_by":1,"name":"Ziduo Yang","email":"","orcid":"","institution":"Jinan University","correspondingAuthor":false,"prefix":"","firstName":"Ziduo","middleName":"","lastName":"Yang","suffix":""},{"id":440313018,"identity":"dd683e2c-47f8-498b-a51b-04c79ec4d15c","order_by":2,"name":"Xiaoqing Liu","email":"","orcid":"","institution":"National university of singapore","correspondingAuthor":false,"prefix":"","firstName":"Xiaoqing","middleName":"","lastName":"Liu","suffix":""},{"id":440313023,"identity":"5f23a54a-3b30-4268-90da-0d49bb0d5d10","order_by":3,"name":"Xiuying Zhang","email":"","orcid":"","institution":"National university of singapore","correspondingAuthor":false,"prefix":"","firstName":"Xiuying","middleName":"","lastName":"Zhang","suffix":""},{"id":440313026,"identity":"c2f00f8a-7810-4d15-b82f-a1ff91bdcde6","order_by":4,"name":"Pengru Huang","email":"","orcid":"","institution":"National university of singapore","correspondingAuthor":false,"prefix":"","firstName":"Pengru","middleName":"","lastName":"Huang","suffix":""},{"id":440313028,"identity":"5f82200c-3a92-4685-a514-cae2994b4309","order_by":5,"name":"Kostya Novoselov","email":"","orcid":"","institution":"National university of singapore","correspondingAuthor":false,"prefix":"","firstName":"Kostya","middleName":"","lastName":"Novoselov","suffix":""}],"badges":[],"createdAt":"2025-03-12 07:53:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6209794/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6209794/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41524-025-01728-w","type":"published","date":"2025-07-15T15:57:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":87219486,"identity":"df2ac6d8-9ffa-4b62-a5f0-8c19031afd98","added_by":"auto","created_at":"2025-07-21 16:05:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6278717,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6209794/v1_covered_3a1c2e6a-71be-4263-a5f4-034741394efb.pdf"},{"id":80735880,"identity":"59cda6c9-d0fe-4b64-828c-4d49106916a5","added_by":"auto","created_at":"2025-04-16 13:27:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":4466212,"visible":true,"origin":"","legend":"","description":"","filename":"SI.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6209794/v1/d5b3d28a10cbc75f114b5a5f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Modeling crystal defects using defect-informed neural networks","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-computational-materials","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjcompumats","sideBox":"Learn more about [npj Computational Materials](http://www.nature.com/npjcompumats/)","snPcode":"41524","submissionUrl":"https://mts-npjcompumats.nature.com/","title":"npj Computational Materials","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Defect Calculations, Materials Discovery, Equivariant Graph Neural Networks, Structural Relaxation","lastPublishedDoi":"10.21203/rs.3.rs-6209794/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6209794/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Machine learning has revolutionized the study of crystalline materials forenabling rapid predictions and discovery. However, most AI-for-Materialsresearch to date has focused on ideal crystals, whereas real-world materialsinevitably contain defects that play a critical role in modern functionaltechnologies. The defects and dopants break geometric symmetry andincrease interaction complexity, posing particular challenges for traditionalML models. Addressing these challenges requires models that areable to capture sparse defect-driven effects in crystals while maintainingadaptability and precision. Here, we introduce Defect-Informed EquivariantGraph Neural Network (DefiNet), a model specifically designedto accurately capture defect-related interactions and geometric configurationsin point-defect structures. Trained on 14,866 defect structures,DefiNet achieves highly accurate structural predictions in a single step,avoiding the time-consuming iterative processes in modern ML relaxationmodels and possible error accumulation from iteration. We furthervalidates DefiNet’s accuracy by using density functional theory (DFT) relaxation on DefiNet-predicted structures. For most defect structures,regardless of defect complexity or system size, only 3 ionic steps arerequired to reach the DFT-level ground state. Finally, comparisonswith scanning transmission electron microscopy (STEM) images confirmDefiNet’s scalability and extrapolation beyond point defects, positioningit as a groundbreaking tool for defect-focused materials research.","manuscriptTitle":"Modeling crystal defects using defect-informed neural networks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-16 13:27:12","doi":"10.21203/rs.3.rs-6209794/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-04-22T11:34:14+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-20T03:37:05+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-18T19:18:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"185622424788879082586787937073919044986","date":"2025-04-08T23:17:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"269374500881638673600804474249496013108","date":"2025-04-02T13:13:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"261935780661685649146406080591130670261","date":"2025-03-28T20:22:03+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-03-28T20:16:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-03-28T16:22:49+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-03-25T07:37:39+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Computational Materials","date":"2025-03-12T07:50:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-computational-materials","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"npjcompumats","sideBox":"Learn more about [npj Computational Materials](http://www.nature.com/npjcompumats/)","snPcode":"41524","submissionUrl":"https://mts-npjcompumats.nature.com/","title":"npj Computational Materials","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4bd547dd-c28e-40c2-8e3b-d346c247e50e","owner":[],"postedDate":"April 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":46877796,"name":"Physical sciences/Materials science/Theory and computation/Atomistic models"},{"id":46877797,"name":"Physical sciences/Physics/Condensed matter physics/Structure of solids and liquids"}],"tags":[],"updatedAt":"2025-07-21T16:04:23+00:00","versionOfRecord":{"articleIdentity":"rs-6209794","link":"https://doi.org/10.1038/s41524-025-01728-w","journal":{"identity":"npj-computational-materials","isVorOnly":false,"title":"npj Computational Materials"},"publishedOn":"2025-07-15 15:57:10","publishedOnDateReadable":"July 15th, 2025"},"versionCreatedAt":"2025-04-16 13:27:12","video":"","vorDoi":"10.1038/s41524-025-01728-w","vorDoiUrl":"https://doi.org/10.1038/s41524-025-01728-w","workflowStages":[]},"version":"v1","identity":"rs-6209794","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6209794","identity":"rs-6209794","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}