High-Throughput Experiments and a Machine-Learning-Driven Analysis to Characterize the Stability of Halide Perovskites

<br/>Machine learning (ML) is a powerful tool to accelerate the development of halide perovskite materials and devices. Because this burgeoning class of material for photovoltaics entails a colossal chemical composition space, ML is very suitable to replace the conventional trial-and-error approach used in their characterization. Thus, there has been a pressing need within the materials research community to identify ML models that can be implemented to inform the physical and chemical behavior of the perovskites. We apply ML models varying from echo state networks to statistical models to classify and predict physical properties such as hole transport layer electrical conductivity, halide perovskite photoluminescence response, the power conversion efficiency of photovoltaic devices, etc. Specifically, we use <i>in situ</i> environmental optical measurements to predict the optical behavior of Cs-FA perovskites for 50+ hours, upon materials’ exposure to moisture. Here, we compare linear regression, echo state network, and seasonal auto-regressive integrated moving average with eXogenous regressor algorithms and attain accuracy of &gt;90% for the latter. Our high-throughput measurements and ML-supported analyses validate the potential of ML to detect and forecast hybrid perovskites’ response with a variety of chemical compositions.