Measurement of information content of Perovskite solar cell’s synthesis descriptors related to performance parameters

Abstract Perovskite solar cells (PSC) are formed by different layers composed of thin films of various materials, in which the properties of every thin layer affect the performance of the cell. The identification of those most relevant properties (or descriptors) has a significant impact on the optimization and cost reduction of the Perovskite solar cell. This relevance is typically evaluated by adjusting a model using subsets of features, but in the present work, we propose to use the mutual information measure to quantify the statistical association between input descriptors and Perovskite solar cell performance parameters ( Voc , Jsc , FF , PCE ). As a result, it is found that ion X is the factor that most impacts the performance of the solar cell. On the other hand, variables such as band gap, Perovskite layer thickness, and A and B ions are also important. In this work, we identify some of the most important factors affecting Perovskite solar cells’ performance, and it could help to improve the efficiency of Perovskite solar cells. In addition, this proposed method could also be applied to other types of functional coatings, thin films, and surfaces.