The elastic, mechanical and optical properties of bismuth modified borate glass: Experimental and artificial neural network simulation

• Published in: Optical materials 2022-04, Vol.126, p.112170, Article 112170
• Main Authors: Effendy, N., Zaid, M.H.M., Sidek, H.A.A., Halimah, M.K., Shabdin, M.K., Yusof, K.A., Mayzan, M.Z.H.
• Format: Article
• Language: English
• Subjects: Artificial neural network; Bismuth borate glass; Mechanical properties; Optical properties
• ISSN: 0925-3467; 1873-1252
• DOI: 10.1016/j.optmat.2022.112170

The introduction of artificial neural networks (ANNs) in the glass field has greatly improved this industry to further increase fabrication productivity. ANNs are the systems that help the glass expert to estimate a few parameters such as density, molar volume, ultrasonic velocity, elastic moduli and optical band gap in the glass composition. The greatness of this system was implemented in a series of bismuth-borate (Bi2O3-B2O3) glasses which have been successfully produced using melting and quenching methods with the configuration of mBi2O3- (100-m)B2O3 where m = 0, 40, 45, 50, 55, 60 mol%. In this present works, the experimental values resulting from the composition of this glass series were compared with the values obtained from the estimation by ANNs. This study has concluded that the ANNs system is relevant to be used in the fields of glass industry since the coefficient of R2 values showed by the density, molar volume, ultrasonic velocity, elastic moduli and optical band gap graph is between 0.998 and 1.0000 which believed highly desirable. •New bismuth-borate glass series were fabricated using melt quenching technique.•The density and molar volume increases with increases of Bi2O3 concentration.•Addition of Bi2O3 improve the elastic properties.•Bond compression and Makishima and Mackenzie simulation support the experimental work.•The coefficient R2 value is between 0.99758 and 1.0000