(257b) Data-Driven Optimization of Perovskite Solar Cells

Perovskite Solar Cells (PSCs) have gained significant attention in the field of photovoltaics in recent years. While lab-scale PSCs have achieved impressive power conversion efficiencies exceeding 25%, there is a lack of systematic and rigorous strategies to optimize various parameters affecting their performance. For instance, material composition and fabrication process variables interact in complex ways, thus presenting a significant challenge for search algorithms. At present, researchers primarily depend on extensive trial-and-error to explore an enormous combinatorial space, which can be both time-consuming and costly, and may lead to suboptimal results.

In this work, we present a combined experimental-computational framework for efficient development of high-performing PSCs. In terms of experimentation, we fabricate organic–inorganic halide perovskite based solar cells via solution processing. We extract photovoltaic parameters from J-V characteristics measured under illumination by a solar simulator. On the computational side, we use black-box optimization methods to guide the search for optimal designs. These algorithms construct models to capture the impact of different factors affecting device performance and predict the desirable combination of design variables. Our results demonstrate that the proposed approach enables rapid and systematic development of high-performing solar devices.