Machine learning (ML)-assisted development of 2D green catalysts to support sustainability

Abstract

Advanced functional two-dimensional (2D) materials have emerged as efficient catalysts for promoting sustainability through the degradation of pollutants and gases. Their tailored features enable diverse catalytic applications, including photocatalysis, piezo-catalysis, and electrocatalysis; however, chemical synthesis of these materials remains a challenge. Therefore, green synthesis of these catalysts is an emerging focus wherein bio-derived and bio-acceptable bioactive catalysts can deal with environmental issues and overcome challenges associated with traditional routes. In this direction, the timely selection and optimization of green catalysts are key factors, requiring exploration through green chemistry and computational analysis. We believe that the involvement of machine learning (ML) in materials science can offer timely catalyst discovery through data-driven predictions and help in developing high-performance catalytic materials required for a sustainable environment. To cover such aspects, this article explores an ML-assisted investigation of efficient, green catalysts via adopting data-driven predictions, thereby assisting in the design and development of a catalyst with desired properties for piezo-catalysis, water splitting, and photocatalysis. This report explores the need for ML to forecast material properties, optimize reaction conditions, and find new catalysts by enhancing computational techniques, such as the density functional theory (DFT), that require a lot of resources. This is a new approach that faces some challenges, which are systematically discussed in this report. The outcomes of this report will serve as guidelines for scholars to explore ML-assisted development of green 2D catalysts, which are needed to achieve high-performance catalysis, thereby managing and maintaining a sustainable environment.