Explainable machine learning for mechanism analysis and slag optimization of boron removal from silicon using slag refining

Explainable machine learning for mechanism analysis and slag optimization of boron removal from silicon using slag refining | 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 Research Article Explainable machine learning for mechanism analysis and slag optimization of boron removal from silicon using slag refining Hong Yue, Guoyu Qian, Zhi Wang, Shijian Dong, Jijun Wu, Yiwei Sun This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8567725/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 07 Apr, 2026 Read the published version in Silicon → Version 1 posted 9 You are reading this latest preprint version Abstract The production of high-purity silicon for advanced applications is critically dependent on the removal of boron, a key impurity notoriously difficult to eliminate. While slag refining is a promising industrial route, its optimization is hindered by complex, multi-parameter interactions that challenge traditional experimental methods. This study employs a machine learning (ML) framework to overcome these limitations. A comprehensive database of 4000 entries was constructed by integrating literature data and thermodynamic calculations. Among six ML algorithms evaluated, an optimized XGBoost model achieved superior predictive accuracy for boron removal efficiency (R² > 0.97, MSE < 0.0003). SHAP analysis identified temperature and the content of the third component (e.g., CaCl₂, CaF₂) as the most influential process parameters. Guided by these insights, an orthogonal experimental design was implemented for optimization. The CaO-SiO₂-CaCl₂ system achieved a peak boron removal efficiency of 88 pct, with the CaO-SiO₂-CaF₂ and CaO-SiO₂-Al₂O₃ systems also reaching high efficiencies of ≥ 86 pct and ≥ 75 pct, respectively. This work establishes a data-driven paradigm that integrates machine learning with metallurgical experimentation for intelligent process optimization. high quality silicon raw materials machine learning silicon purification slag refining boron removal Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 07 Apr, 2026 Read the published version in Silicon → Version 1 posted Editorial decision: Revision requested 27 Feb, 2026 Reviews received at journal 15 Feb, 2026 Reviews received at journal 13 Feb, 2026 Reviewers agreed at journal 07 Feb, 2026 Reviewers agreed at journal 05 Feb, 2026 Reviewers invited by journal 05 Feb, 2026 Editor assigned by journal 16 Jan, 2026 Submission checks completed at journal 16 Jan, 2026 First submitted to journal 10 Jan, 2026 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-8567725","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":587437544,"identity":"e9ac7f0a-931e-4352-a90a-6b76398f731a","order_by":0,"name":"Hong Yue","email":"","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Hong","middleName":"","lastName":"Yue","suffix":""},{"id":587437545,"identity":"6b4703a4-efb9-4d3e-829d-4e3529c78401","order_by":1,"name":"Guoyu Qian","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuklEQVRIiWNgGAWjYJCCAwwMNjwka0kjUQsQHCZBrXxE8sYDP3eclzFnb2D88IPBLo+gFsMbaQUHe8/c5rHsOcAs2cOQXExYy4wcgwO8bbd5DG4kMEgD/ZXYQIyWg3/bzvEY3H/A/JsoLfISOQaHedsOAG1hYCPOFgOeZwWHZduSgX5JbLPsMUgmwpb25M0f37bZ2ZuzHz5840eFHRG2XEgwgDAYGBtAJGEg338ApmUUjIJRMApGAQ4AAF4NPgGTqg46AAAAAElFTkSuQmCC","orcid":"","institution":"Institute of Process Engineering","correspondingAuthor":true,"prefix":"","firstName":"Guoyu","middleName":"","lastName":"Qian","suffix":""},{"id":587437546,"identity":"a2d75c1c-0b74-4748-84f0-14e5d5b9dfc2","order_by":2,"name":"Zhi Wang","email":"","orcid":"","institution":"Institute of Process Engineering","correspondingAuthor":false,"prefix":"","firstName":"Zhi","middleName":"","lastName":"Wang","suffix":""},{"id":587437547,"identity":"d9078d29-0594-47ea-8454-1127a5279ffe","order_by":3,"name":"Shijian Dong","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Shijian","middleName":"","lastName":"Dong","suffix":""},{"id":587437548,"identity":"c8b5f835-35be-4f78-b6d0-a964b8677d55","order_by":4,"name":"Jijun Wu","email":"","orcid":"","institution":"Kunming University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Jijun","middleName":"","lastName":"Wu","suffix":""},{"id":587437551,"identity":"ce2e0173-85bb-4267-9442-436b1bb84cdb","order_by":5,"name":"Yiwei Sun","email":"","orcid":"","institution":"Institute of Process Engineering","correspondingAuthor":false,"prefix":"","firstName":"Yiwei","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2026-01-10 10:53:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8567725/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8567725/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s12633-026-03681-9","type":"published","date":"2026-04-07T15:57:24+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":106809079,"identity":"38965018-93ba-49ad-8922-b9c7e3c14feb","added_by":"auto","created_at":"2026-04-13 16:06:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1601929,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8567725/v1_covered_0aa41e2f-8f71-4995-9da0-1ec5b5b71808.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Explainable machine learning for mechanism analysis and slag optimization of boron removal from silicon using slag refining","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":"silicon","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scon","sideBox":"Learn more about [Silicon](https://www.springer.com/journal/12633)","snPcode":"12633","submissionUrl":"https://submission.nature.com/new-submission/12633/3","title":"Silicon","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"high quality silicon raw materials, machine learning, silicon purification, slag refining, boron removal","lastPublishedDoi":"10.21203/rs.3.rs-8567725/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8567725/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe production of high-purity silicon for advanced applications is critically dependent on the removal of boron, a key impurity notoriously difficult to eliminate. While slag refining is a promising industrial route, its optimization is hindered by complex, multi-parameter interactions that challenge traditional experimental methods. This study employs a machine learning (ML) framework to overcome these limitations. A comprehensive database of 4000 entries was constructed by integrating literature data and thermodynamic calculations. Among six ML algorithms evaluated, an optimized XGBoost model achieved superior predictive accuracy for boron removal efficiency (R\u0026sup2; \u0026gt; 0.97, MSE\u0026thinsp;\u0026lt;\u0026thinsp;0.0003). SHAP analysis identified temperature and the content of the third component (e.g., CaCl₂, CaF₂) as the most influential process parameters. Guided by these insights, an orthogonal experimental design was implemented for optimization. The CaO-SiO₂-CaCl₂ system achieved a peak boron removal efficiency of 88 pct, with the CaO-SiO₂-CaF₂ and CaO-SiO₂-Al₂O₃ systems also reaching high efficiencies of \u0026ge;\u0026thinsp;86 pct and \u0026ge;\u0026thinsp;75 pct, respectively. This work establishes a data-driven paradigm that integrates machine learning with metallurgical experimentation for intelligent process optimization.\u003c/p\u003e","manuscriptTitle":"Explainable machine learning for mechanism analysis and slag optimization of boron removal from silicon using slag refining","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-10 06:43:59","doi":"10.21203/rs.3.rs-8567725/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-27T14:27:38+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-15T08:21:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-13T09:56:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"79615718102313176165379494846306008121","date":"2026-02-07T12:32:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"44814564523041364949239069478772410971","date":"2026-02-05T16:12:01+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-05T12:30:54+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-16T08:00:36+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-16T07:59:54+00:00","index":"","fulltext":""},{"type":"submitted","content":"Silicon","date":"2026-01-10T10:36:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"silicon","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scon","sideBox":"Learn more about [Silicon](https://www.springer.com/journal/12633)","snPcode":"12633","submissionUrl":"https://submission.nature.com/new-submission/12633/3","title":"Silicon","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"779d7904-3806-4c38-8894-41dbe2ee902f","owner":[],"postedDate":"February 10th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-13T16:01:59+00:00","versionOfRecord":{"articleIdentity":"rs-8567725","link":"https://doi.org/10.1007/s12633-026-03681-9","journal":{"identity":"silicon","isVorOnly":false,"title":"Silicon"},"publishedOn":"2026-04-07 15:57:24","publishedOnDateReadable":"April 7th, 2026"},"versionCreatedAt":"2026-02-10 06:43:59","video":"","vorDoi":"10.1007/s12633-026-03681-9","vorDoiUrl":"https://doi.org/10.1007/s12633-026-03681-9","workflowStages":[]},"version":"v1","identity":"rs-8567725","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8567725","identity":"rs-8567725","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}