Despite its promising performance for C2+ product formation in electrochemical reduction of CO2 (CO2RR), Cu-based catalysts exhibit a relatively poor selectivity towards important chemical intermediates such as ethylene and ethanol. Up to 16 different products can typically be formed during CO2RR.1 Cu is the only element that can form C-C bonds at an appreciable rate.2 Therefore, it is critical to better understand the active sites of Cu-based electrocatalysts, as it can lead to better catalyst design with improved selectivity towards desired products. A common approach is to investigate the reactivity of well-defined surface terminations of Cu, which can be achieved by employing shape-controlled (nano)particles.3 Here, we expanded on such studies by evaluating the impact of Au on well-defined Cu2O surfaces for the electrochemical reduction of CO2. This promoter metal was chosen, because it can reduce CO2 to CO, which is widely accepted as the key intermediate for C-C coupling.2,4 In this study, 40 nm Cu2O nanocubes that predominantly expose the (100) facet were synthesized with a ligand-free method. The (100) facet is more active in the production of ethylene at the expense of unwanted methane formation, which takes place predominantly at the (111) facet.3 The surface of the nanocubes was decorated with Au nanoparticles by a galvanic replacement reaction. HAADF-STEM images of the catalysts with various Au loadings are shown in Figure 1a-d. After addition of Au nanoparticles, the shape of the Cu2O nanocubes was preserved. However, the Au nanoparticles were highly dispersed at low loadings (1 mol % Au) whereas Au clusters were observed at higher loadings (10 mol % Au). This result was in line with synchrotron X-ray diffraction (XRD) results, whereby the Au(111) reflection appears for the 10Au/Cu2O sample. Besides, the formation of a Au-Cu alloy and CuO phase was confirmed as well. The highly dispersed nature of the Au nanoparticles at low loadings was further established by Au 4f X-ray photoelectron spectroscopy (XPS), whereby a shift towards higher binding energies was observed as compared to the samples with higher Au loadings. The electrocatalytic performance of the catalysts was assessed over a range of potentials in neutral electrolyte (0.1 M KHCO3) in H-cell configuration. With highly dispersed Au as promotor, we found that the formation of ethylene and ethanol was significantly enhanced (Figure 1e-f). More specifically, the Faradaic effiency towards ethanol increased by roughly 1.5 times by introducing 1 mol % Au to the Cu2O nanocubes. Surprisingly, the 10Au/Cu2O sample displayed the lowest Faradaic efficiency for C2 products. We believe that the dispersion of Au on the Cu2O surface plays an important role in the product formation, whereby CO can be formed at Au sites and diffuse to Cu sites in proximity for further reduction to C2 products (CO spillover). To further study structure-activity relationships, we performed in-situ X-ray absorption spectroscopy (XAS) measurements. Figure 1g shows the normalized Cu K-edge X-ray absorption near-edge structure (XANES) of the Cu2O and Au/Cu2O samples in the final state. The evolution of the Cu(0), Cu(I) and Cu(II) fractions was followed by Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) analysis. This analysis revealed that the final oxidation state of the catalysts is different. The Cu(I)/Cu(0) ratio shows the following trend 1Au/Cu2O > 5Au/Cu2O > Cu2O > 10Au/Cu2O. Based on these observations, we hypothesize that the oxidation state of Cu affects the product distribution and in particular, the formation of oxygenates. The presence of Cu(I) species and the formation of Cu(0) species during CO2 electroreduction was further confirmed with in-situ XRD measurements. Finally, we performed quasi in-situ XPS measurements to study the reduction of the surface of the catalysts. Acknowledgements This publication is part of the research programme 'Reversible Large-scale Energy Storage' (RELEASE) with project number 17621 which is financed by the Dutch Research Council (NWO).