Predicting the mass attenuation coefficient of glass systems using machine learning approach for radiation applications

In this study, we present a machine learning approach to predict the mass attenuation coefficients (MAC) of some glassy materials, aTeO2-bMoO3-cBaO-dSm2O3-eBi2O3, across various weight fractions (a, b, c, d, e) and energy levels between 0.015 and 15 MeV energy range. The compound exhibits a diverse set of weight fractions, ensuring a comprehensive representation of possible compositions. The NIST XCOM database serves as the reference for scattering and coherent values. This study introduces a novel and potentially alternative method for determining MAC in glass systems, with potential extensions to evaluate diverse radiation shielding characteristics. In contrast to traditional software, this approach presents unique advantages. The utilization of this technique holds promise for designing innovative glass compositions tailored for specific radiation shielding applications.