The electrochemical pathway of CO₂ reduction reaction (CO₂RR) remains a promising technique to convert CO₂ to value-added chemicals (e.g. CO, CH₄, HCOOH, CH₃OH, etc), which helps alleviate our dependence on fossil fuels as well as mitigate the rising CO₂ concentration in the atmosphere. Nowadays, most of the CO₂RR study focus on designing novel catalysts to improve their activity and selectivity. However, to evaluate the performance of new catalysts, a long time electrolysis (e.g. 30 min to several hours) is usually required to build up the concentration of products (both gaseous and liquid) enough before product analysis is performed by gas chromatography-mass spectrometry (GCMS) or nuclear magnetic resonance (NMR). This results in a long “pre-concentration-collection-detection” cycle and it greatly limits the efficiency of catalyst screening process as well as increases the energy cost. In addition, for catalysts that degrade quickly with time (within minutes), it is nearly impossible to monitor their product distribution change using traditional detection methods, which directly leads to the product information loss during the study of their degradation mechanisms. Therefore, it is urgent to develop a novel and sensitive detection method that can perform product analysis in a very short time scale (e.g. within minutes).

In this work, we present a systematical study of using RRDE as a fast and reliable detection method for quantifying CO₂RR products generated from both heterogeneous (e.g. Pt, Au, Sn) and homogeneous (e.g. [Ni^II(cyclam)]²⁺) catalysts. Local pH change during Au and Sn catalyzed CO₂RR are also carefully investigated. The fast detection towards single product systems (e.g. H₂, CO) and mixture product systems (e.g. H₂ and HCOO^-, H₂ and CO) are discussed in detail. Based on Hori’s classification, catalysis systems of Pt, Au and Sn encompass nearly 90% of the heterogeneous CO₂RR reduction products. Our results show that, compared with traditional detection methods (e.g. GCMS, NMR), RRDE shows a superior detection sensitivity (e.g. LOD_ring ~10^-18 moles of H₂, defined in this work) as well as a significant short detection time (<1 min). Meanwhile, the product distribution as a function of applied electrolysis potential could be easily extracted from the detected ring voltammograms, which provides important information for the evaluation of catalysts’ performance.