Machine-learning-guided high-throughput design of asymmetric A–DA′D–A acceptors toward efficient organic solar cells

Key Points

• This research aims to develop a strategy for designing efficient asymmetric acceptors for organic solar cells using machine learning.
• Utilized machine learning algorithms to predict the performance of molecular candidates.
• Conducted high-throughput virtual screening of over one million NFAs (non-fullerene acceptors).
• Matched new asymmetric acceptors with multiple donor molecules.
• Identified several new asymmetric acceptors that show promise for enhancing solar cell efficiency.
• Achieved improved performance metrics compared to previous acceptor designs.