Revealing the Untapped Potential of Photocatalytic Overall Water Splitting in Metal Organic Frameworks

Abstract

The past decade witnessed substantial attention toward metal‐organic frameworks (MOFs) for photocatalytic water splitting owing to their versatile structural and optoelectronic characteristics. However, MOFs capable of efficient photocatalytic overall water splitting (OWS) remain notably scarce. Although MOF‐based photocatalysts with OWS potential are highly promising due to their diverse building blocks and topological configurations, the vast number of possible MOFs renders traditional trial‐and‐error materials discovery approaches impractical. Herein, a data‐driven methodology that integrates machine learning with high‐throughput first‐principles computations to identify MOFs with OWS capability is presented. By systematically assessing factors including water stability, band gap, band positions, charge carrier transport, and optical absorption properties, 14 MOFs from the Quantum‐MOF (QMOF) database containing over 20,000 MOFs as promising candidates for visible‐light‐driven OWS are identified. Notably, five of them exhibit exceptional electronic and optical properties, outperforming previously reported MOF OWS photocatalysts, such as UIO66(Zr)‐NH2, MIL125(Ti)‐NH2, and MIL53(Al)‐NH2 is established. This work represents a large‐scale, data‐driven exploration of MOF‐based photocatalysts for water splitting, shedding light on the untapped potential of photocatalysis in MOFs.