Universal machine learning aided synthesis approach of two-dimensional perovskites in a typical laboratory

This paper proposes a universal framework integrating high-throughput experiments, chemical knowledge, and machine learning (ML) techniques to guide the synthesis of 2D silver/bismuth organic-inorganic hybrid perovskites. The framework uses subgroup discovery and support vector machines to uncover structure-property relationships and screen for materials with high synthesis feasibility. Applying this approach increased the synthesis success rate fourfold compared to traditional methods, offering a practical route for accelerating material synthesis with limited resources.

Yilei Wu, Chang-Feng Wang, Ming-Gang Ju, Qiangqiang Jia, Qionghua Zhou, Shuaihua Lu, Xinying Gao, Yi Zhang, Jinlan Wang