Towards efficient and stable organic solar cells: fixing the morphology problem in block copolymer active layers with synergistic strategies supported by interpretable machine learning†

Achieving outstanding photovoltaic performance in terms of power conversion efficiency (PCE) and long-term stability establishes the basis for commercial application of organic solar cells (OSCs). However, OSCs’ development universally faces a contradiction from these two aspects. To address this critical challenge, we take a morphologically stable donor–acceptor block copolymer (BCP) and optimize its morphology using two types of small-molecule additives to increase the PCE. The suppressed acceptor block crystallinity and the disturbed electron transport pathway in the neat BCP are the targets in this study. Benefiting from calculation-guided experimental design, we discover an unexpected synergistic optimization between the morphological and electrical tuning realized by the two types of additives, one of which acts as an n-type dopant. The latter strengthens the non-covalent attraction between the BCP acceptor blocks to repair the BCP morphology; meanwhile, the other small-molecule acceptor helps to reduce the doping reaction energy barrier to enhance the doping effect. With the aid of interpretable machine learning, we confirm the structured correlation between the morphology, the electrical parameters, and the ultimate photovoltaic performance. The synergistic optimization enhances the PCE from 13.2% to 15.9% with excellent 83% PCE maintenance after 85 °C aging for 1000 h. This impressive combination encourages further OSC development without a traditional compromise between the PCE and thermal stress lifetime.