Energy differences as descriptors for the correlation between and in nonfullerene organic photovoltaics.

ITIC-series nonfullerene organic photovoltaics (NF OPVs) have realized the simultaneous increases of the short-circuit current density () and open-circuit voltage (), called the positive correlation between and , which could improve the power conversion efficiency (PCE). However, it is complicated to predict the formation of positive correlation in devices through simple calculations of single molecules due to their dimensional differences. Here, a series of symmetrical NF acceptors blended with the PBDB-T donor were chosen to establish an association framework between the molecular modification strategy and positive correlation. It can be found that the positive correlation is modification site-dependent following the energy variation at the different levels. Furthermore, to illustrate a positive correlation, the energy gap differences (Δ) and the energy level differences of the lowest unoccupied molecular orbitals (Δ) between the two changed acceptors were proposed as two molecular descriptors. Combined with the machine learning model, the accuracy of the proposed descriptor is more than 70% for predicting the correlation, which verifies the reliability of the prediction model. This work establishes the relative relationship between two molecular descriptors with different molecular modification sites and realizes the prediction of the trend of efficiency. Therefore, future research should focus on the simultaneous enhancement of photovoltaic parameters for high-performance NF OPVs.