https://doi.org/10.1140/epjs/s11734-025-01572-5
Regular Article
Integrating machine learning and experimental data in modeling optical behaviors of neodymium oxide nanoparticle-doped glasses
1
Institute of Informatics and Computing in Energy (IICE), College of Engineering, Universiti Tenaga Nasional, 43000, Kajang, Selangor, Malaysia
2
Department of Electronic Systems Engineering (ESE), Malaysia–Japan International Institute of Technology (MJIIT), Universiti Teknologi Malaysia, Johor Bahru, Malaysia
3
Centre for Pre-University Studies, Universiti Malaysia Sarawak, 94300, Kota Samarahan, Sarawak, Malaysia
4
Department of Chemistry, College of Science, King Khalid University, 61413, Abha, Saudi Arabia
a nazirulnazrin@ymail.com, nazirul.nazrin@uniten.edu.my
Received:
2
September
2024
Accepted:
16
February
2025
Published online:
29
March
2025
Zinc tellurite doped with neodymium oxide nanoparticles (Nd2O3 NPs) glass series has been fabricated using conventional melt-quenching technique to explore the respective optical, structural, and physical properties of the glass material. Fourier-transform infra-red (FTIR) spectra and the respective deconvolution revealed changes in the amount of TeO3 and TeO4 structural units related to the concentration of neodymium oxide nanoparticles, which directly influences the amount of bridging oxygen and non-bridging oxygen in the glass system. The density of the prepared glasses possesses increasing trend with values that rose from 5346 to 5606 kg m−3 as more Nd2O3 NPs are incorporated in the glass matrix. The parameters for optical properties have also been determined, assisted and trained the chosen model in order to predict the data for optical properties of glass material with different chemical compositions via the machine learning approach. To enhance the experimental results, machine learning models were developed using linear regression (LR) and artificial neural networks (ANN). ANN model with its ability to capture both linear as well as non-linear interactions has outperformed the simulation and LR model as the ANN model had managed to accurately predict the values for the optical properties of glass material based on the compositional parameters. This underscores the effectiveness of ANN in modeling having intricate relationships between dopant concentrations and optical behaviors while at the same time providing critical insights for future advancements in doped glass technology.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025
corrected publication 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
