Continuous Electrolysis of Carbon Dioxide for the Production of Formate with High Efficiency

Carbon dioxide capture, storage, and utilization is nowadays considered one of the most interesting approaches to reduce emissions to the atmosphere and, therefore, to mitigate climate change. In this sense, the electrochemical reduction of carbon dioxide to value-added products is gaining attention, since it permits storing excess of energy from renewable and intermittent sources (e.g. solar or wind), in the form of chemicals [1]. Among the reaction products, both formic acid and formate are products of great interest that could be obtained by this process, due to its use as raw materials in several industries and its promising use as a reactant in low-temperature fuel cells and as a renewable hydrogen carrier molecule. Although there is a large number of studies published in literature focusing on the electrocatalytic reduction of carbon dioxide to formic acid and formate[2], few of them achieve relevant trade-offs among the figures of merit typically used to analyze the performance of the process, like (i) the concentration of formate, (ii) the Faradaic Efficiency towards the reaction products, (iii) the energy consumption per kmol of formate and (iv) the current density supplied to the electrochemical reactor. In this context, this communication aims at studying the continuous electrolysis of carbon dioxide, operating with a single pass of the reactants through the electrochemical reactor and using Bi-carbon supported electrocatalysts, to obtain formate with high energy efficiency. In this sense, the electrocatalysts are deposited by air-brushing techniques in the form of a membrane electrode assembly in the cathodic compartment of the electrochemical reactor. This configuration permits direct contact between the Gas Diffusion Electrode and the Nafion cationic exchange membrane which acts as a separator between both compartments of the electrochemical filter press reactor [3]. The influence of some key variables is rigorously studied, such as (i) the amount of water in the carbon dioxide input stream, (ii) the current density and (iii) the Bi catalyst loading deposited over the cathode. Firstly, the performance of the electrocatalytic reduction of carbon dioxide to formate employing a Bi catalyst loading of 0.75 mg·cm -2 in terms of i) formate concentration, ii) Faradaic Efficiency towards formate and iii) energy consumption per kmol of formate are assessed as a function of the current density and the water flow in the carbon dioxide input stream. In this point, it is important to remark that, working with an industrially-relevant current density of 200 mA·cm -2 and with a water flow of 0.5 g·h -1 , high formateconcentrations of up to 312 g·L -1 were obtained with Faradaic Efficiency towards formateof 25 %, respectively. Consequently, additional tests are performed with a Bi loading in the cathode of the electrochemical filter press of 1.5 mg·cm -2 . Combinations of relevant values of concentrations, Faradaic Efficiency for formate and energy consumption of 337 g·L -1 , 89 %, and 180 kWh·kmol -1 , respectively, are achieved with a current density of 45 mA·cm -2 and with a water flow of 2 g·h -1 . These results are one of the best trade-offs reported in the field of carbon dioxide electroreduction to formic acid or formate among these figures of merit, thus achieving a significant advance in this field. However, there are many challenges to be addressed for the implementation of the process at an industrial scale, and therefore, more research focused on the advanced electrocatalytic materials, innovative electrode configuration, and improved electrochemical reactors are still required. References [1] A. Irabien, M. Alvarez-Guerra, J. Albo, A. Dominguez-Ramos, Electrochem. Water Wastewater Treat. Elsevier, 2018 , pp. 29–59. [2] X. Zhang, S. X. Guo, K. A. Gandionco, A. M. Bond, J. Zhang, Mater. Today Adv., 2020 , 7, 100074. [3] G. Díaz-Sainz, M. Alvarez-Guerra, J. Solla-Gullón, L. García-Cruz, V. Montiel, A. Irabien, Chem. Eng. J., 2021 , 405, 126965. Acknowledgments The authors acknowledge the financial support from the Spanish State Research Agency (AEI) through the projects PID2019-108136RB-C31, and PID2020-112845RB-I00 (AEI/10.13039/501100011033). Guillermo Díaz Sainz would like to show their gratitude to the financial support from the European Regional Development Fund in the framework of the Interreg Atlantic program through the project “HYLANTIC”-EAPA_204/2016”.