Predicting Band Gaps of 2D Phosphorene-Isoelectronic materials Using Machine Learning

Abstract

Two-dimensional (2D) materials have emerged as promising candidates for next-generation electronic and optoelectronic devices due to their unique structural and electronic properties. Among these materials, phosphorene and its isoelectronic counterparts have attracted considerable attention because of their tunable bandgaps and excellent charge transport characteristics. However, conventional bandgap calculations based on Density Functional Theory (DFT) require substantial computational resources and become inefficient when screening large material datasets. This study presents a machine learning-based framework for predicting the bandgaps of 2D phosphorene-isoelectronic materials. A dataset containing structural, electronic, and physicochemical descriptors of various 2D materials was utilized to train and evaluate multiple regression models, including Random Forest Regression, Support Vector Regression, and Gradient Boosting Regression. Feature selection techniques were employed to identify the most influential parameters affecting bandgap behavior and to improve model efficiency. The performance of the developed models was assessed using standard evaluation metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and the coefficient of determination (R²). The obtained results indicate that machine learning algorithms can accurately estimate bandgap values while significantly reducing computational cost compared with traditional first-principles calculations. The proposed approach highlights the potential of data-driven methodologies for accelerating the discovery and design of novel 2D semiconducting materials with tailored electronic properties.