Data-Driven Discovery of Robust Materials for Photocatalytic Energy Conversion

Abstract

The solar-to-chemical energy conversion of Earth-abundant resources like water or greenhouse gas pollutants like CO2 promises an alternate energy source that is clean, renewable, and environmentally friendly. The eventual large-scale application of such photo-based energy conversion devices can be realized through the discovery of novel photocatalytic materials that are efficient, selective, and robust. In the past decade, the Materials Genome Initiative has led to a major leap in the development of materials databases, both computational and experimental. Hundreds of photocatalysts have recently been discovered for various chemical reactions, such as water splitting and carbon dioxide reduction, employing these databases and/or data informatics, machine learning, and high-throughput computational and experimental methods. In this article, we review these data-driven photocatalyst discoveries, emphasizing the methods and techniques developed in the last few years to determine the (photo)electrochemical stability of photocatalysts, leading to the discovery of photocatalysts that remain robust and durable under operational conditions.