![Author ORCID: We display the ORCID iD icon alongside authors names on our website to acknowledge that the ORCiD has been authenticated when entered by the user. To view the users ORCiD record click the icon. [opens in a new tab]](https://chemrxiv.org/engage/assets/public/chemrxiv/images/logos/orcid.png)
Abstract
Organic-inorganic perovskite solar cells (PSCs) are promising candidates for next-generation, inexpensive solar panels due to their high power conversion efficiency, which is on par with their commercial silicon counterparts. However, PSCs suffer from poor stability. A new subset of PSCs, quasi-two-dimensional Ruddlesden-Popper PSCs (quasi-2D RP PSCs), is known for improved photostability and superior resilience to environmental conditions in comparison with three-dimensional (3D) metal-halide PSCs. To expedite the search of new quasi-2D RP PSCs we report a combinatorial, machine learning (ML) enhanced high-throughput perovskite film fabrication and optimization study. We designed a bespoke experiment strategy and produced perovskite films with a range of different compositions through a fully automated drop-casting process. The performance and characterization data of these solar cells were used to train a ML model that allowed for material parameter optimization and directed the design of improved materials. The ML optimized quasi-2D RP perovskite films yielded solar cells with power conversion efficiencies reaching 16.3%.
