Abstract
Understanding the relationships between molecular properties and device parameters is highly desired not only to improve the overall performance of an organic solar cell but also to fulfill the requirements of a device for a particular application such as solar to fuel energy conversion (high Voc) and solar-window applications (high Jsc). In this work, a series of machine learning models are built for three important device characteristics (Voc, Jsc, and FF) using 13 crucial molecular properties as descriptors, with an impressive predictive performance (r=0.7). These models may play a vital role in designing promising organic materials for a specific photovoltaic application with high Voc/Jsc. The importance of descriptors for each device parameter is unraveled, which may assist in tuning them and improve understanding of the energy conversion process.
Funding Information
  • Ministry of Education of the People's Republic of China (020414380126)
  • National Natural Science Foundation of China (21673109, 21722302)

This publication has 61 references indexed in Scilit: