Instructional Design Based on First-Principles Calculations and Machine Learning:Simulation and Prediction of Band Gap of Bismuth-Based Materials
TANG Ran-Xiao1, ZHANG Heng2, WEI Qiu-Hong1**, MA Ying2**, ZHAO Ying1, LI Hui-Liang1, CHEN Xiao-Cui1
1. College of Science,Hebei Agricultural University,Baoding 071001,China; 2. School of Chemistry and Chemical Engineering,Shandong University,Jinan 250100,China
Abstract First-principles calculations and machine learning,as essential tools in contemporary computational chemistry and materials science,have played pivotal roles in scientific research.Introducing them into undergraduate experimental teaching is of great significance.Traditional experimental teaching on obtaining material band gaps often relies on experimental measurements or commercial software calculations,lacking a systematic integration of data-driven methods.To address this,this project focuses on predicting the band gaps of bismuth-based materials and designs a digital experimental workflow that integrates first-principles calculations with machine learning,utilizing completely open-source and free software platforms.Relying on the Matminer toolkit and the SNUMAT database,458 bismuth-based materials were screened,and 145-dimensional features were extracted.Various machine learning models were constructed using the Orange platform to predict the HSE band gaps.Concurrently,the PBE band structure of BiOBr was calculated using Quantum ESPRESSO,thereby establishing a collaborative computational pathway of“theoretical simulation+data-driven modeling”.This design highlights cross-disciplinary integration and cost-effective instructional innovation.It not only helps students master the entire workflow from structure optimization and band structure calculation to machine learning modeling,but also strengthens their scientific modeling and data analysis capabilities.The pedagogical outcomes demonstrate that this project possesses strong generalizability and practical application value.
TANG Ran-Xiao, ZHANG Heng, WEI Qiu-Hong, MA Ying, ZHAO Ying, LI Hui-Liang, CHEN Xiao-Cui. Instructional Design Based on First-Principles Calculations and Machine Learning:Simulation and Prediction of Band Gap of Bismuth-Based Materials[J]. Chinese Journal of Chemical Education, 2026, 47(12): 57-64.
