Yan Ge, Tang Hongcai, Shen Yangzi, Han Liyuan, Han Qifeng
State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Shanghai, 200240, China.
Adv Mater. 2025 Jul;37(26):e2503154. doi: 10.1002/adma.202503154. Epub 2025 May 6.
The 2D/3D heterojunction perovskite solar cells (PSCs) exhibit remarkable stability, but the quantum well in the 2D perovskite capping layer hinders the carrier transport, thereby lowering the power conversion efficiency (PCE). The relationship between the transport barrier and the complex structure of ammonium ligands (ALs) is currently poorly understood, thus leading to the one-sided approach and inefficient process in the development of 2D perovskite. Here, a machine learning procedure is established to comprehensively explore the relationship and combined it with an artificial intelligence (AI) model based on reinforcement learning algorithm to accelerate the generation of ALs. Finally, the AI-designed ALs improved the carrier transport performance of the 2D perovskite capping layer, and we achieved a certified PCE of 26.12% in inverted PSCs. The devices retained 96.79% of the initial PCE after 2000 h operation in maximum power point tracking under 1-sun illumination at 85°C.
