Electrochemical Reduction of CO2 to Multi-Carbon Products on Sputtered Cu Nanoparticles Gas Diffusion Electrodes

The electrochemical reduction of CO 2 is a technology of great environmental and economic interest as it provides a promising solution for reducing the concentration of CO 2 with producing valuable chemical compounds. Cu and its alloys have attracted much attention because of their moderate CO binding energy for reducing CO 2 to multi-carbon (C2+) products[1]. The reactivity and product selectivity of the CO 2 reduction reaction (CO 2 RR) depend on the interfacial structure and the nano-scale morphology of the catalyst as well as on the catalyst material composition. Especially for the gas diffusion electrodes (GDEs), the catalyst morphology has a significant influence on the formation of triple-phase boundary that increases CO 2 reactivity and product selectivity[2,3]. However, the influence of catalyst morphology including catalyst thickness, porosity, particle size on CO 2 RR has not been fully understood yet because of the complex CO 2 RR mechanism and catalyst structure in GDEs. In this study, we investigated the relationship between product selectivity and the thickness of the Cu controlled by sputtering techniques and characterized nanostructure of the Cu-GDEs. The Cu catalyst layer (CL) was deposited on a commercial carbon-based gas diffusion layer (GDL) with a micro porous layer (MPL) by magnetron sputtering from a pure Cu target at room temperature. The geometric area of the deposited Cu was 0.5 cm 2 . The nanostructures of GDEs were evaluated using scanning transmission electron microscopy equipped with energy dispersive X-ray analyzer (STEM-EDX). Electrochemical experiments were carried out in a three-electrode setup using a Biologic-VSP instrument. A home-made three-compartment flow cell was used with the Cu-GDE as the working electrode between the gas and cathode compartments. An Ag/AgCl electrode was placed in the cathode compartment as the reference electrode and a Pt mesh as the counter electrode in the anode compartment. 1 M KCl was used as the catholyte, saturated KHCO 3 was used as the anolyte and a proton exchange membrane (Nafion 117) was used to separate the anode and cathode compartments. The CO 2 gas flow to a gas compartment was kept at 30 sccm. Electrochemical CO 2 RR was performed by chronopotentiometry. The gas products were quantified using a gas chromatography with thermal conductivity detector. As for the liquid products, alcohols were evaluated using a gas chromatography with flame ionization detector and organic acids were detected by high performance liquid chromatography with conductivity detector, respectively. The faradaic efficiencies (FEs) of CO 2 RR products using GDEs with various CL thickness of 70, 300, 1000, 2000 nm were measured under current density of 400 mA cm -2 . The GDE with CL thickness of x nm is written as CLxGDE. The major gas and liquid products for all the GDEs were ethylene and ethanol with the FE of around 37-42% and 26-31%, respectively. As decreasing CL thickness from 2000 nm, FEs for C2+ products (FE C2+ ), such as ethylene, ethanol, n-propanol etc., were increased from that of CL2000GDE of 69%. The maximum FE C2+ reaching 81% was achieved for CL300GDE. The FE C2+ for CL70GDE was slightly decreased down to 76%. In comparison with the FEs for CL2000GDE, the FEs of ethylene, ethanol, and n-propanol were each increased by 2-5%, while FE of H 2 was decreased for CL300GDE. To clarify the details of CL morphology, the surface structure of CL300GDEs was observed before and after the CO 2 RR measurement using STEM-EDX. Figures (a) and (b) show cross-sectional images of nano-scale annular dark field (ADF)-STEM and EDX mapping (purple color show Cu and cyan blue color show C) before the CO 2 RR measurement. The Cu layer was uniformly stacked on MPL with the designed thickness of 300 nm. Figures (c) and (d) show cross-sectional images of ADF-STEM and EDX mapping after the CO 2 RR measurement. Cu nanoparticles less than 50 nm were distributed on the surface and inside of the MPL. The stacked Cu layer as observed before the CO 2 RR measurement was not observed after the CO 2 RR. These results indicate that Cu atoms migrate during the CO 2 RR and distributed Cu nanoparticles exhibit the high FE C2+ . We will discuss the influence of Cu migration during the CO 2 RR on product selectivity from the results of synchrotron-based analysis and nanostructure observation in the presentation. [1] A. Bagger et al., ChemPhysChem. 2017, 18, 3266-3273. [2] N. T. Nesbitt et al., ACS Catal. 2020, 10, 14093-14106. [3] A. Inoue et al., EES Catal. 2023, 1, 9–16. Fig. Cross-sectional images of (a), (c) ADF-STEM and (b), (d) EDX mapping before and after CO 2 RR for CL300GDEs, respectively. Purple color show Cu and cyan blue color show C. Figure 1