Yayın:
Structure-Informed Design of Float-Compatible Thin Glass Supported by Data-Driven Screening

dc.contributor.authorGosterislioglu, Yekta Ates
dc.contributor.authorArslan, Banu
dc.contributor.authorSokmen, Ilkay
dc.contributor.authorCelikbilek Ersundu, Miray
dc.contributor.authorErsundu, Ali Ercin
dc.date.accessioned2026-06-27T15:36:54Z
dc.date.issued2026
dc.description.abstractMotivated by growing demand in electronics, automotive, photovoltaic, and architectural applications, thin-glass products offer substantial potential for material savings and reduced environmental impact, yet their widespread adoption is limited by mechanical instability and reliance on non-float manufacturing routes. In this work, we establish a comprehensive framework for the design of mechanically robust thin-glass compositions compatible with conventional float-glass production. The approach combines (i) structure-informed, composition-structure-property-driven design and (ii) data-driven screening based on artificial intelligence-assisted predictive modeling. While the data-driven approach is used to efficiently explore a broad compositional space and identify candidate glasses with high intrinsic elastic modulus and hardness, the primary optimization of chemical strengthening response and processability is achieved through structure-informed design. Glass compositions within the SiO2-Al2O3-B2O3-RO-R ' 2O system (R = Mg, Ca; R ' = Li, Na, K) were designed to target chemical temperability, mechanical durability, and float-line processability, and were evaluated through thermal, optical, structural, and mechanical analyses. The optimized compositions achieved stress layer depths up to similar to 40 & micro;m after chemical tempering, enabling high mechanical performance at reduced thickness. The best-performing glasses exhibited a similar to 10% increase in fracture load and a similar to 25% increase in surface hardness at a thickness of 0.75 mm. When benchmarked against standard 1.6 mm soda-lime-silica glass, this corresponds to a reduced areal weight of 1.85 kg m-2 (similar to 50% lower mass) and an estimated similar to 60% reduction in process-related CO2 emissions per square meter, based on reduced material usage within the defined boundary. By integrating data-driven screening with structure-informed composition design, this work establishes a scalable pathway for next-generation thin flat glass development, supporting environmentally sustainable and energy-efficient glass manufacturing.en
dc.description.sponsorshipScientific and Technological Research Council of Trkiye (TBIdot
dc.description.sponsorshipTAK) [119C098]
dc.description.urihttps://doi.org/10.1111/jace.70885
dc.identifier.doi10.1111/jace.70885
dc.identifier.eissn1551-2916
dc.identifier.issn0002-7820
dc.identifier.issue6
dc.identifier.urihttps://hdl.handle.net/20.500.14981/72027
dc.identifier.volume109
dc.identifier.wos001777662800001
dc.language.isoeng
dc.publisherWILEY
dc.relation.ispartofJOURNAL OF THE AMERICAN CERAMIC SOCIETY
dc.rightsopenAccess
dc.subjectartificial intelligence-assisted predictive modeling
dc.subjectchemical strengthening
dc.subjectcomposition-structure-property relationships
dc.subjectsustainable glass manufacturing
dc.subjectthin flat glass
dc.subjectALKALI ALUMINOSILICATE GLASSES
dc.subjectPROPERTY RELATIONSHIPS
dc.subjectELASTIC PROPERTIES
dc.subjectSTRENGTH
dc.subjectRESISTANCE
dc.subjectSTATE
dc.subjectMaterials Science
dc.titleStructure-Informed Design of Float-Compatible Thin Glass Supported by Data-Driven Screening
dc.typeArticle
dspace.entity.typePublication
local.import.sourceWOS

Dosyalar

Koleksiyonlar