Utilization of Machine Learning to Map-Out Features of Potential Perovskite Solar Cell Compositions

Project Description

Project Overview:  Perovskites structured compound, most commonly a hybrid organic-inorganic lead or tin halide-based material, are being investigated as an alternative to silicon solar cells. They have unique electronic properties that have the potential to be more efficient than existing solar cells. By altering the composition of a perovskite crystal structure, the properties of the material change.   This project will use machine learning and Density Functional Theory (DFT) generated data to fit a model with the goal of finding an ideal composition subspace with promising stability and bandgap. The team will use machine learning mythology to improve experimental error bounds.  The team will use the Materials Simulation Toolkit for Machine Learning (MAST-ML) to create machine learning model fits.

Client: Dr. Dane Morgan, University of Wisconsin – Madison, Department of Materials Science and Engineering

Student Team: Raman Gill, Dana Katz, Joseph Kern, Avery Ulschmid