MXene-Integrated Perovskite Tandem Solar Cells for Sustainable Development Goals (SDGs): A Bibliometric and Physics-Informed Artificial Intelligence (AI) Analysis Using Graph Neural Networks and Transformer Models

Mihaira H. Haddad (Prince Sattam bin Abdulaziz University)
Fatima Hassan Mohammed (University of Khartoum)
Khadersab Adamsab (University of Technology and Applied Sciences–Al Musannah)

ASEAN Journal for Science and Engineering in Materials
Vol 5, No 3 (2026): AJSEM: Volume 5, Issue 3, December 2026

This study presents a bibliometric and artificial intelligence (AI) analysis of MXene-integrated perovskite tandem solar cells to support renewable energy innovation and the Sustainable Development Goals (SDGs). A literature-based bibliometric perspective was used to identify research trends in perovskite tandem photovoltaics, MXene materials, and AI-assisted solar cell optimization. A physics-informed computational framework integrating Graph Neural Networks (GNNs) and Transformer models was then developed using 200 photovoltaic samples with 26 material, environmental, and performance variables. The GNN model captured nonlinear material–performance relationships, while the Transformer model predicted retention and degradation behavior under humidity and thermal stress. Feature importance analysis identified MXene loading, conductivity, defect density, and interface quality as dominant factors affecting efficiency and stability.