Machine-learning-assisted exploration of new non-fullerene acceptors for high-efficiency organic solar cells

• DOI: 10.1016/j.xcrp.2024.102316
• Publication Date: 2024-12-11T15:55:33Z
• Authors: Zhikang Zhu; Chenyang Zhu; Yibo Tu; Tianxiang Shao; Yida Wang; Weihong Liu; Yiming Liu; Yue Zang; Qingya Wei; Wensheng Yan
• Keywords: Overfitting

Highlights•Comprehensive database of training models•Explainable models are used to predict the PCE organic solar cells•Three non-fullerene acceptor molecules were designed based on SHAP analysis•The error between model prediction and experimental verification is less than 3%SummaryThe power conversion efficiency (PCE) cells (OSCs) has exceeded 19% with development acceptors (NFAs). Here, machine learning (ML) inputs both molecular descriptors fingerprints different algorithms investigated assist exploration NFAs. Although exhibits slightly inferior performance parameters, it can deliver faster high-throughput computation much stronger generalization ability due decreased complexity avoid overfitting. Moreover, Shapley additive explanations (SHAP) techniques explain for design synthesis three An excellent agreement predicted PCEs achieved, a relative 3%. Therefore, our study offer strategy rapid detailed analysis screening NFAs high-efficiency OSCs.Graphical abstract