CO₂ Reduction beyond Copper-Based Catalysts: A Natural Language Processing Review from the Scientific Literature

Abstract

Carbon dioxide (CO 2 ) is a prominent greenhouse gas that contributes significantly to global warming. To combat this issue, one strategy is the conversion of CO 2 into alcohols and hydrocarbons, which can be used as fuels and chemical feedstocks. Consequently, a substantial volume of scientific literature has been dedicated to investigating different materials and reaction conditions to facilitate the CO 2 reduction reaction (CO 2 RR) into these so-called high-value products. However, the vastness of this literature makes it challenging to stay updated on recent discoveries and review the most promising materials and conditions that have been explored. To address this issue, we applied natural language processing tools to extract valuable data from 7292 published articles in the scientific literature. Our analysis revealed the emergence of new materials such as cesium−lead−bromide perovskites and bismuth oxyhalides that have been recently used in the CO 2 RR and identified Bi-based catalysts as the most selective for HCOO − production. Furthermore, we gleaned insights into the composition of other elements and materials commonly employed in the CO 2 RR, their relationship to product distribution, and the prevalent electrolytes used in the CO 2 electrochemical reduction. Our findings can serve as a foundation for future investigations in the realm of catalysts for CO 2 RRs, offering insights into the most promising materials and conditions to pursue further research.