Data-Driven Discovery of a Covalent Organic Framework Heterojunction as Efficient Photocatalysts for Overall Solar Water Splitting

Searching for highly efficient visible-light photocatalysts is a high-cost and time-consuming process in the water splitting field. The integration of data-driven screening based on the database and density functional theory calculations represents a promising approach. In this study, we first present a topologically assembled single-layer covalent organic framework (COF) that is used to build a COF heterojunction database via AA stacking. Then we propose a systematic search procedure for COF heterojunctions as overall solar water splitting photocatalysts, including suitable band gap (screen 1), appropriate band edge position (screen 2), spontaneous catalytic reactions for water splitting (screen 3), and efficient separation of photogenerated electrons and holes (proof). Finally, we successfully identify 1 heterojunction from the pool of 222 items as an efficient photocatalyst for overall solar water splitting. Clearly, this kind of data-driven screening procedure, based on a COF heterojunction database, opens up new avenues and inspires the development of high-performance photocatalysts.