{"paper_id":"26f38920-5f54-4b3d-b5cc-0cfa9f82c21e","body_text":"Hydrogen Diffusion in Magnesium Using Machine Learning Potentials | 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 Hydrogen Diffusion in Magnesium Using Machine Learning Potentials Andrea Angeletti, Luca Leoni, Dario Massa, Luca Pasquini, Stefanos Papanikolaou, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4773688/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Mar, 2025 Read the published version in npj Computational Materials → Version 1 posted 12 You are reading this latest preprint version Abstract Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice. These interactions span large length and time scales, making them difficult to address with standard ab initio techniques. This work addresses this challenge by employing accelerated machine learning (ML) molecular dynamics simulations through active learning. We conduct a comparative study of different ML-based interatomic potential schemes, including VASP, MACE, and CHGNet, utilizing various training strategies such as on-the-fly learning, pre-trained universal models, and fine-tuning. We obtain an optimal hydrogen diffusion coefficient value of 2.1 x 10 -8 m 2 s -1 at 673 K in MgH 0.06 , which aligns exceptionally well with experimental results, underlining the efficacy and accuracy of ML-assisted methodologies in the context of diffusive dynamics. Particularly, our procedure significantly reduces the computational effort associated with traditional transition state calculations or ad-hoc designed interatomic potentials. The results highlight the limitations of pre-trained universal solutions for defective materials and how they can be improved by fine-tuning. Specifically, fine-tuning the models on a database produced during on-the-fly training of VASP ML force field allows the retrieving of DFT-level accuracy at a fraction of the computational cost. Physical sciences/Physics/Condensed-matter physics/Electronic properties and materials Physical sciences/Materials science/Materials for energy and catalysis Full Text Additional Declarations (Not answered) Supplementary Files Supplementary.pdf Supplementary Materials for ”Hydrogen Diffusion in Magnesium Using Machine Learning Potentials” Supplementary.pdf Cite Share Download PDF Status: Published Journal Publication published 31 Mar, 2025 Read the published version in npj Computational Materials → Version 1 posted Editorial decision: revise 29 Sep, 2024 Review # 3 received at journal 20 Sep, 2024 Review # 2 received at journal 03 Sep, 2024 Review # 1 received at journal 28 Aug, 2024 Reviewer # 3 agreed at journal 21 Aug, 2024 Reviewer # 2 agreed at journal 21 Aug, 2024 Reviewer # 1 agreed at journal 15 Aug, 2024 Reviewers invited by journal 15 Aug, 2024 Submission checks completed at journal 25 Jul, 2024 First submitted to journal 22 Jul, 2024 Unknown event 22 Jul, 2024 Editor assigned by journal 20 Jul, 2024 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-4773688\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":340829507,\"identity\":\"23ef9a86-f71c-48dd-8415-aba980e0e3de\",\"order_by\":0,\"name\":\"Andrea Angeletti\",\"email\":\"data:image/png;base64,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\",\"orcid\":\"https://orcid.org/0000-0002-7142-6768\",\"institution\":\"University of Vienna, Vienna Doctoral School in Physics\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Andrea\",\"middleName\":\"\",\"lastName\":\"Angeletti\",\"suffix\":\"\"},{\"id\":340829508,\"identity\":\"f56944cc-63ad-4459-81e7-4e58e043c2d1\",\"order_by\":1,\"name\":\"Luca Leoni\",\"email\":\"\",\"orcid\":\"https://orcid.org/0009-0007-5648-0719\",\"institution\":\"Alma Mater Studiorum - Università di Bologna\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Luca\",\"middleName\":\"\",\"lastName\":\"Leoni\",\"suffix\":\"\"},{\"id\":340829509,\"identity\":\"607ba47c-88d5-40de-b2a1-74b6a3d8ba4a\",\"order_by\":2,\"name\":\"Dario Massa\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-3131-3996\",\"institution\":\"NCBJ\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Dario\",\"middleName\":\"\",\"lastName\":\"Massa\",\"suffix\":\"\"},{\"id\":340829510,\"identity\":\"1bddf23d-649e-487c-88ef-93e6accc5936\",\"order_by\":3,\"name\":\"Luca Pasquini\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Alma Mater Studiorum - Università di Bologna\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Luca\",\"middleName\":\"\",\"lastName\":\"Pasquini\",\"suffix\":\"\"},{\"id\":340829511,\"identity\":\"3f52dc48-f892-4970-881b-54b034570684\",\"order_by\":4,\"name\":\"Stefanos Papanikolaou\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Centre of Nuclear Research\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Stefanos\",\"middleName\":\"\",\"lastName\":\"Papanikolaou\",\"suffix\":\"\"},{\"id\":340829512,\"identity\":\"57c9bf15-1594-45cd-b443-6c09e554edaa\",\"order_by\":5,\"name\":\"Cesare Franchini\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-7990-2984\",\"institution\":\"University of Vienna\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Cesare\",\"middleName\":\"\",\"lastName\":\"Franchini\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-07-20 15:05:06\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-4773688/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-4773688/v1\",\"draftVersion\":[],\"editorialEvents\":[{\"content\":\"https://doi.org/10.1038/s41524-025-01555-z\",\"type\":\"published\",\"date\":\"2025-03-31T04:00:00+00:00\"}],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":79553801,\"identity\":\"68f896b6-c940-4d3e-9c78-e594085460c6\",\"added_by\":\"auto\",\"created_at\":\"2025-03-31 07:05:48\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1661292,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"ANGELETTInpjCompMat.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4773688/v1_covered_dad24abb-ee52-4d60-a879-92757699357b.pdf\"},{\"id\":64357407,\"identity\":\"1141d695-adce-4856-bf41-cc849f7a4272\",\"added_by\":\"auto\",\"created_at\":\"2024-09-12 06:13:54\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":706634,\"visible\":true,\"origin\":\"\",\"legend\":\"Supplementary Materials for \\u0026#x201D;Hydrogen Diffusion in Magnesium Using Machine Learning Potentials\\u0026#x201D;\",\"description\":\"\",\"filename\":\"Supplementary.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4773688/v1/b9644060dc37c022ce670b8e.pdf\"},{\"id\":64357408,\"identity\":\"0a6bbb95-f407-41e5-9bce-d44f20ebf185\",\"added_by\":\"auto\",\"created_at\":\"2024-09-12 06:13:54\",\"extension\":\"pdf\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":706634,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Supplementary.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-4773688/v1/1ead757cd87a95aca87e5563.pdf\"}],\"financialInterests\":\"(Not answered)\",\"formattedTitle\":\"Hydrogen Diffusion in Magnesium Using Machine Learning Potentials\",\"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\":\"info@researchsquare.com\",\"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\":\"\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-4773688/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-4773688/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice. These interactions span large length and time scales, making them difficult to address with standard ab initio techniques. This work addresses this challenge by employing accelerated machine learning (ML) molecular dynamics simulations through active learning. We conduct a comparative study of different ML-based interatomic potential schemes, including VASP, MACE, and CHGNet, utilizing various training strategies such as on-the-fly learning, pre-trained universal models, and fine-tuning. We obtain an optimal hydrogen diffusion coefficient value of 2.1 x 10\\u003csup\\u003e-8\\u003c/sup\\u003e m\\u003csup\\u003e2\\u003c/sup\\u003es\\u003csup\\u003e-1\\u003c/sup\\u003e at 673 K in MgH\\u003csub\\u003e0.06\\u003c/sub\\u003e, which aligns exceptionally well with experimental results, underlining the efficacy and accuracy of ML-assisted methodologies in the context of diffusive dynamics. Particularly, our procedure significantly reduces the computational effort associated with traditional transition state calculations or ad-hoc designed interatomic potentials. The results highlight the limitations of pre-trained universal solutions for defective materials and how they can be improved by fine-tuning. Specifically, fine-tuning the models on a database produced during on-the-fly training of VASP ML force field allows the retrieving of DFT-level accuracy at a fraction of the computational cost.\",\"manuscriptTitle\":\"Hydrogen Diffusion in Magnesium Using Machine Learning Potentials\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-09-12 06:13:49\",\"doi\":\"10.21203/rs.3.rs-4773688/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"revise\",\"date\":\"2024-09-30T02:39:38+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"This content is not available.\",\"date\":\"2024-09-20T15:40:35+00:00\",\"index\":3,\"fulltext\":\"This content is not available.\"},{\"type\":\"editorInvitedReview\",\"content\":\"This content is not available.\",\"date\":\"2024-09-03T22:55:42+00:00\",\"index\":2,\"fulltext\":\"This content is not available.\"},{\"type\":\"editorInvitedReview\",\"content\":\"This content is not available.\",\"date\":\"2024-08-28T21:36:40+00:00\",\"index\":1,\"fulltext\":\"This content is not available.\"},{\"type\":\"reviewerAgreed\",\"content\":\"This content is not available.\",\"date\":\"2024-08-21T08:59:26+00:00\",\"index\":3,\"fulltext\":\"This content is not available.\"},{\"type\":\"reviewerAgreed\",\"content\":\"This content is not available.\",\"date\":\"2024-08-21T05:29:57+00:00\",\"index\":2,\"fulltext\":\"This content is not available.\"},{\"type\":\"reviewerAgreed\",\"content\":\"This content is not available.\",\"date\":\"2024-08-15T23:29:26+00:00\",\"index\":1,\"fulltext\":\"This content is not available.\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2024-08-15T20:08:09+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2024-07-25T14:57:32+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"npj Computational Materials\",\"date\":\"2024-07-22T16:33:03+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksFailed\",\"content\":\"\",\"date\":\"2024-07-22T14:41:29+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2024-07-20T15:00:29+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"1974ec09-29af-4569-a8e9-e8c02612f308\",\"owner\":[],\"postedDate\":\"September 12th, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"published-in-journal\",\"subjectAreas\":[{\"id\":36112893,\"name\":\"Physical sciences/Physics/Condensed-matter physics/Electronic properties and materials\"},{\"id\":36112894,\"name\":\"Physical sciences/Materials science/Materials for energy and catalysis\"}],\"tags\":[],\"updatedAt\":\"2025-03-31T07:05:41+00:00\",\"versionOfRecord\":{\"articleIdentity\":\"rs-4773688\",\"link\":\"https://doi.org/10.1038/s41524-025-01555-z\",\"journal\":{\"identity\":\"npj-computational-materials\",\"isVorOnly\":false,\"title\":\"npj Computational Materials\"},\"publishedOn\":\"2025-03-31 04:00:00\",\"publishedOnDateReadable\":\"March 31st, 2025\"},\"versionCreatedAt\":\"2024-09-12 06:13:49\",\"video\":\"\",\"vorDoi\":\"10.1038/s41524-025-01555-z\",\"vorDoiUrl\":\"https://doi.org/10.1038/s41524-025-01555-z\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-4773688\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-4773688\",\"identity\":\"rs-4773688\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}