An Extrinsic Faradaic Layer on CuSn for High-Performance Electrocatalytic CO 2 Reduction

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 May 2022An Extrinsic Faradaic Layer on CuSn for High-Performance Electrocatalytic CO2 Reduction Feilong Ren, Wenjian Hu, Cheng Wang, Pin Wang, Wenbo Li, Congping Wu, Yingfang Yao, Wenjun Luo and Zhigang Zou Feilong Ren Eco-Materials and Renewable Energy Research Center (ERERC), Jiangsu Key Laboratory for Nano Technology, National Laboratory of Solid State Microstructures, Department of Physics, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Wenjian Hu National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Cheng Wang National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Pin Wang Eco-Materials and Renewable Energy Research Center (ERERC), Jiangsu Key Laboratory for Nano Technology, National Laboratory of Solid State Microstructures, Department of Physics, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Wenbo Li National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Congping Wu Eco-Materials and Renewable Energy Research Center (ERERC), Jiangsu Key Laboratory for Nano Technology, National Laboratory of Solid State Microstructures, Department of Physics, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Yingfang Yao National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093 Google Scholar More articles by this author , Wenjun Luo *Corresponding author: E-mail Address: [email protected] National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093 Google Scholar More articles by this author and Zhigang Zou Eco-Materials and Renewable Energy Research Center (ERERC), Jiangsu Key Laboratory for Nano Technology, National Laboratory of Solid State Microstructures, Department of Physics, Nanjing University, Nanjing 210093 National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202100794 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail An intrinsic Faradaic layer on the surface of a metal electrocatalyst is usually considered an active site for CO2 reduction. Different strategies have been used to improve the performance of CO2 reduction by adjusting the intrinsic Faradaic layer. However, it is still challenging to achieve CO2 reduction with high activity, selectivity, and stability. In this study, for the first time, we improve the three parameters simultaneously by introducing a Zn(OH)x over layer onto a CuSn electrocatalyst. We find that the intrinsic Faradaic layer of Sn(OH)x on the surface of CuSn provides active sites for CO2 reduction, while Zn(OH)x plays multiple roles as an adsorption/activation layer, a cover layer, and a protective layer. Further studies suggest that the enhanced activity comes from a Faradaic reaction of Zn(OH)x during CO2 reduction, which can be considered as an extrinsic Faradaic layer. This new strategy of introducing an extrinsic Faradaic layer can deepen understanding of electrocatalytic process and offers guidance to design other high-performance electrocatalysts. Download figure Download PowerPoint Introduction Electrochemical reduction of carbon dioxide (CO2) by renewable energy, such as solar or wind, attracts wide interest because it is such a promising way to solve the CO2 emission problem.1–3 In previous studies, different metals have been used as electrocatalysts for CO2 reduction.4–6 Generally, there are intrinsic hydroxyl layers on the surfaces of metal electrocatalysts7–10 on which Faradaic reactions (coupled electron and ion reactions) occur and provide active sites for CO2 reduction during electrocatalysis.11–16 Therefore, the surface hydroxyl layers can be considered as intrinsic Faradaic layers. Some strategies have been used to improve the selectivity and activity of CO2 reduction by modifying the intrinsic Faradaic layers of electrocatalysts.17 For example, the Faradaic efficiency (FE) and activity for formic acid on Sn metal catalysts were improved by increasing the amount of the surface hydroxyl layer.18 Moreover, by regulating the thickness of the intrinsic Faradaic layer, a Ag–Sn bimetallic catalyst exhibited higher FE and CO2 reduction activity.19 The selectivity and activity on the surface hydroxyl layer were also enhanced by stabilizing reduction intermediates.7,18,19 Though different methods have been used to modify the intrinsic Faradaic layers, the efficiency of electrocatalysts is still not high. For example, metallic Sn is often exposed on the surface of the electrocatalyst, which tends to reduce H+ into hydrogen and lowers the FE of CO2 reduction.7 Moreover, the intrinsic Faradaic layers are usually unstable during electrocatalytic CO2 reduction, which leads to attenuated FE.12,20–22 To date, electrocatalysts with high activity, selectivity, and stability have not been achieved, which limits their practical applications. Therefore, it is desirable to develop a facile strategy to simultaneously improve these three parameters of electrocatalysts for CO2 reduction. In this study, using CuSn metal as a model electrocatalyst, we simultaneously improved the activity, selectivity, and stability of the electrocatalyst for CO2 reduction by introducing a Zn(OH)x overlayer. Zn(OH)x was selected to combine with CuSn due to its good stability at negative potentials (see Supporting Information Table S1). Moreover, the reduction potential of Zn(OH)x is close to the onset potential of CuSn (about −0.7 VRHE) for formic acid production.23,24 We found that except for the intrinsic Faradaic layer of Sn(OH)x, the Faradaic reaction also occurs on Zn(OH)x during CO2 reduction. Therefore, Zn(OH)x can be considered as an extrinsic Faradaic layer. Moreover, although challenging, it is very important to identify the function of different components in composite catalysts.25–27 Therefore, we also studied the roles of different Faradaic layers during CO2 reduction in detail. The results suggest that Sn(OH)x, not Zn(OH)x, provides the active sites. Zn(OH)x increases the adsorption of CO2 and spills over intermediates onto Sn(OH)x to produce formic acid, which can improve the activity of CO2 reduction remarkably. Moreover, Zn(OH)x can cover the exposed CuSn metal completely and increase the selectivity and stability of CO2 reduction. Experimental Preparation of samples CuSn and CuSn/Zn(OH)x were prepared on carbon paper by electroplating and subsequently by electrooxidation. Sodium citrate (14.705 g) was dissolved into 100 mL deionized water by stirring at room temperature. CuCl2·2H2O (0.34 g), SnCl2 (0.255 g), and ZnCl2 (2.726 g) were successively added into the solution to obtain blue transparent precursor solution. The electrodeposition was employed in a three-electrode cell with an electrochemical analyzer (CHI760E; ChenHua, Shanghai, China). A carbon paper, a carbon rod, and an Ag/AgCl electrode were used as a work electrode, a counter electrode, and a reference electrode, respectively. First, the CuSn and CuSnZn films were electrodeposited on carbon paper by cyclic voltammetry (CV) at the potential range of −1.1 to −1.2 V (vs Ag/AgCl). The scan rate was 10 mV/s. The deposition charges for CuSn and CuSnZn were both 3C. After electrodeposition, the CuSn and CuSnZn films were washed by deionized water and then dried in air. Zn film, as a reference sample, was also deposited on carbon paper at the potential range of −1.4 to −1.5 V (vs Ag/AgCl) in 100 mL sodium citrate aqueous solution (14.705 g) with ZnCl2 (2.726 g). Second, the as-deposited CuSn, CuSnZn, and Zn films were electrooxidized to CuSn/Sn(OH)x, CuSn/Sn(OH)x/Zn(OH)x, and Zn(OH)x as electrocatalysts, respectively. The electrooxidation process was carried out in a H-type cell separated by a proton-exchange membrane (NafionN117, DuPont, Wilmington, DE, USA). A Pt sheet and an Ag/AgCl electrode were used as a counter electrode and a reference electrode, respectively. The three samples were electrooxidized through CV at the potential range of 0.5 to −2.0 V (vs Ag/AgCl) for 20 cycles in CO2-saturated 0.5 M KHCO3 solution. The scan rate was 100 mV/s. Finally, the three electrodes after electrooxidation were washed by deionized water and dried in air. Characterization of samples The surface morphologies and the surface element ratios of the samples were investigated by scanning electron microscopy (SEM; Nano Nova 230, FEI, Hillsboro, OR, USA) and energy-dispersive X-ray spectroscopy (SEM-EDX; Genesis XM2 60S, Ametek, Santiago, CA, USA), respectively. The crystal structures of the samples were investigated by transmission electron microscopy (TEM; FEI Tecnai F20, 200 kV, FEI, Hillsboro, OR, USA) and X-ray diffraction (XRD; smartlab, 9KW, Rigaku, Matsubaracho, Akishima, Tokyo, Japan). The element distribution of the samples was investigated by energy-dispersive X-ray element mapping (EDX-mapping; FEI Tecnai F20, 200kV, FEI, Hillsboro, OR, USA). The surface compositions and the chemical states were analyzed by X-ray photoelectron spectroscopy (XPS; PHI5000 VersaProbe, ULVAC-PHI, Chichibu, Saitama, Japan) with Al Kα radiation (1486.6 eV). The binding energies were calibrated with C1s at 284.6 eV. The CO2 adsorption of samples was analyzed by the Brunauer–Emmer–Teller (BET; TriStar 3000, Micromeritics, Atlanta, GA, USA) method at 273 K. Electrocatalytic CO2 reduction measurement Electrochemical measurements were carried out in a three-electrode cell with an electrochemical analyzer (CHI760E) at room temperature. An as-prepared sample, a Pt sheet, and an Ag/AgCl electrode were used as a work electrode, a counter electrode, and a reference electrode, respectively. A H-type cell was separated by a Nation 117 membrane. A work electrode and a reference electrode were put into a cathodic compartment, and a Pt counter electrode was put into an anodic compartment. The electrolyte was 0.5 M KHCO3 aqueous solution. Before measurement, CO2 gas was bubbled into the electrolyte for 30 min to obtain CO2-saturated solution (pH 7.2). During measurement, CO2 was passed into the cathodic compartment at a rate of 10 mL/min, and the electrolyte was stirred slowly. The FEs were tested at different potentials for 30 min. After the CO2 reduction reaction, 0.1 mL electrolyte was collected and mixed with 0.1 mL p-toluenesulfonic acid. A high-performance liquid chromatography (HPLC; Shimadzu, Japan) was used to analyze formic acid in the mixture solution. Other reduction products, H2 and CO, were detected by gas chromatography (GC-8A and GC-2014; Shimadzu, Japan). The FEs of the products were calculated by the following formula: FE ( product ) = z × n × F Q where z is the number of electron transfers needed for CO2 reduction, n is the total amount of the final products, F is the Faraday constant (96,485 C/mol), and Q is the total amount of charge passed through the sample. The partial current density was calculated by the following formula: j HCOOH = j total × FE HCOOH where jtotal is the total current density and FEHCOOH is the FE for HCOOH. Results and Discussion A CuSn/Zn(OH)x sample was prepared on carbon paper by electroplating28 and subsequently by electoroxidation. As a reference sample, CuSn was also prepared by the same method. Figures 1a and 1b and Supporting Information Figure S1 indicate the SEM images of the CuSn and CuSn/Zn(OH)x. The two samples were both particle aggregates and indicated similar morphology. The compositions of CuSn and CuSn/Zn(OH)x were confirmed by TEM (see Figures 1c and 1d) and XPS ( Supporting Information Figures S2 and S3). A lattice distance of 2.11 Å was observed on both of CuSn and CuSn/Zn(OH)x, which was assigned as the (660) plane of Cu81Sn22. The XRD peaks at 42.63° and 42.65° were observed on both of the two samples, which were assigned as the (660) plane of intermetallic Cu81Sn22 (42.638°) and were in agreement with the TEM results (see Supporting Information Figure S4). From the XPS spectra, the chemical states of Cu on the surface of CuSn were the same as those on CuSn/Zn(OH)x, which were both composed of Cu0 and Cu2+. However, the chemical states of Sn were different on the two samples. Both Sn0 and Sn (II/IV) were observed on the surface of the CuSn sample, but only Sn (II/IV) on the surface of CuSn/Zn(OH)x. Sn0 and Sn (II/IV) on the surface of a CuSn sample possibly came from intermetallic CuSn and Sn(OH)x, respectively. In CuSn/Zn(OH)x, the binding energies of Zn2p and O1s were assigned as Zn2+ and lattice OH−, respectively.29,30 To further analyze the compositions of the two samples, element distributions of CuSn and CuSn/Zn(OH)x were examined by EDX-mapping (see Figures 1e and 1f and Supporting Information Figure S5). In the two samples, the distribution profiles of metal element were similar to the central region of the particle. Moreover, the morphologies and surface chemical states of Zn(OH)x reference sample were also characterized by SEM and XPS and the results are shown in Figures S5 and S6. Figure 1 | SEM images of (a) CuSn and (b) CuSn/Zn(OH)x. TEM images of (c) CuSn and (d) CuSn/Zn(OH)x. EDX-mapping images of (e) CuSn and (f) CuSn/Zn(OH)x. Download figure Download PowerPoint The FE values and current densities on CuSn and CuSn/Zn(OH)x were measured in CO2-saturated 0.5 M KHCO3 solution three times. The average values and standard deviation are shown in Supporting Information Figures S8 and S9. Figure 2a indicates the average FE on CuSn and CuSn/Zn(OH)x. The same reduction products, HCOOH and H2, as well as very little CO, were obtained on both of the two samples. At a more positive reduction potential, H2 was a major product on the two samples. The FE for H2 decreased, but the efficiency for HCOOH increased with increasing reduction potentials. At a more negative reduction potential, HCOOH was a major product on CuSn and CuSn/Zn(OH)x due to the higher energy barrier for CO2 reduction than for H2 reduction.31 CuSn/Zn(OH)x indicated a higher FE for HCOOH than that of CuSn at the potential range from −0.6 to −1.0 VRHE. In particular, we found that the FE enhancement depended on the applied potentials’ sensitivity, which was different from CuSn and indicated a peak value at the potential range of −0.7 to −0.8 VRHE. A very high FE (90%) for HCOOH was obtained on CuSn/Zn(OH)x at −1.0 VRHE, while a FE for H2 was only about 5% at this potential. To eliminate the effect of Cu, we also measured the FE for HCOOH on Sn with and without Zn(OH)x. The results are shown in Supporting Information Figure S10. Similar to CuSn/Zn(OH)x, Sn/Zn(OH)x still indicated a higher FE for HCOOH than Sn at the potential range from −0.7 to −1.0 VRHE. Consequently, Zn(OH)x improved the CO2RR performance on both CuSn and Sn. Moreover, CuSn/Zn(OH)x indicated much higher FE for HCOOH than Sn/Zn(OH)x. Therefore, we investigated the CO2RR properties of CuSn/Zn(OH)x in this study. Figure 2b indicates the average partial current of formic acid on CuSn and CuSn/Zn(OH)x, with Zn(OH)x as a reference sample. Compared with CuSn and CuSn/Zn(OH)x, the partial current of formic acid on Zn(OH)x was negligible at the potential range of −0.6 to −0.9 VRHE. Therefore, not Zn(OH)x but CuSn was the active site for CO2 reduction. The average partial current density of CuSn/Zn(OH)x was about 12 mA/cm2 at −1.0 VRHE, much higher than the sum of CuSn (6.8 mA/cm2) and Zn(OH)x (2 mA/cm2) at the same potential. To investigate the reasons for improved current density, we measured the electrochemical surface area (ECSA) of CuSn/Zn(OH)x, CuSn, and Zn(OH)x, and the results are shown in Supporting Information Figures S11–S13. The ECSA of CuSn/Zn(OH)x was about 1.5 times as high as that of CuSn, and Zn(OH)x indicating much lower ECSA than CuSn/Zn(OH)x and CuSn. However, the average partial current density for HCOOH on CuSn/Zn(OH)x was about three times as high as that of CuSn at −0.7 VRHE. Therefore, the enhanced current density came from not only higher surface area, but also other factors. According to the above results, the selectivity and activity on CuSn were significantly improved by introducing Zn(OH)x layers. Figure 2 | (a) FEs for H2, CO, and formic acid on CuSn and CuSn/Zn(OH)x. (b) Partial current densities of formic acid on Zn(OH)x, CuSn, and CuSn/Zn(OH)x. SEM images of CuSn/Zn(OH)x before (c) and after etching (d) by 0.1 M KOH. (e) XPS spectra of Sn 3d in CuSn/Zn(OH)x before and after etching. (f) CV curves of CuSn/(OH)x before and after etching, CuSn and Zn(OH)x as references, electrolyte: CO2-saturated 0.5 M KHCO3 aqueous solution (pH 7.2). Download figure Download PowerPoint To investigate the role of the Zn(OH)x layer during electrocatalysis CO2 reduction, CuSn/Zn(OH)x after 3 h’s stability measurement was etched by 0.1 M KOH aqueous solution for 30 min, which can remove Zn(OH)x selectively. The ratio of Zn/Cu decreased from 0.77 to 0.1 after etching (see Supporting Information Table S2), which suggested that most of the Zn(OH)x was removed but a little was still left in the composite sample. The SEM images of CuSn/Zn(OH)x before and after etching are also shown in Figures 2c and 2d, respectively. Dendrite Zn(OH)x was observed on the surface of CuSn/Zn(OH)x. After etching, nanoparticle CuSn was exposed, which was very similar to the as-prepared CuSn sample (see Figure 1a). Moreover, XPS was used to characterize the surface change of CuSn/Zn(OH)x before and after etching. Before etching, only one peak at 485.9 eV was observed on the CuSn/Zn(OH)x, which was assigned as Sn4+ (see Figure 2e). However, one new peak at 484.5 eV of Sn0 appeared after etching,7 which suggested that Zn(OH)x covers the surface Sn0 in CuSn/Zn(OH)x. The results are also in good agreement with Supporting Information Figures S2 and S3. The FE and partial current density for formic acid on CuSn/Zn(OH)x before and after etching were measured and the results are shown in Supporting Information Figure S14. CuSn/Zn(OH)x after etching indicated much lower FE and partial current density than those of the sample before etching, which confirms that Zn(OH)x layer can remarkably improve the selectivity and activity of CO2 reduction on CuSn. According to previous studies,7,20 a hydrogen evolution reaction occurs on Sn0 in CuSn electrocatalyst, which is a key competition reaction with CO2 reduction. Herein, Sn0 was covered completely on CuSn by Zn(OH)x and suppressed the H2 evolution reaction, which led to higher selectivity for HCOOH. To further confirm the role of the Zn(OH)x layer during electrocatalysis CO2 reduction, we measured the FE for HCOOH of CuSn/Zn(OH)x with a different amount of Zn(OH)x at −0.9 VRHE in 0.5 M KHCO3 solution, and the results are shown in Supporting Information Figure S15. The amount of surface Zn(OH)x was controlled by adjusting the concentration of ZnCl2 in the electrolyte during the process of electrodepositing CuSn/Zn(OH)x. The relation between FE for HCOOH and the amount of Zn(OH)x indicated a volcanic type curve, and the highest FE was obtained at the concentration of 200 mM of ZnCl2, which further confirmed that the surface Zn(OH)x can improve the FE for HCOOH on CuSn. Moreover, to understand the mechanism of the activity improvement by introducing the Zn(OH)x layer, the CV curves of CuSn/Zn(OH)x before and after etching, with CuSn and Zn(OH)x as references, were measured, and the results are indicated in Figure 2f. Four reduction peaks at 0.30, −0.46, −0.81, and −1.07 VRHE were observed on CuSn/Zn(OH)x before etching, respectively. The peaks at 0.30 and −0.46 VRHE were also observed on the CuSn sample, while the peak at −0.81 VRHE only occurred on Zn(OH)x. The three reduction peaks at 0.30, −0.46, and −0.81 VRHE were assigned as redox peaks of Cu+/Cu, Sn2+/Sn, and Zn2+/Zn, respectively.11,32,33 Different from the three reduction peaks, the shoulder reduction peak at −1.07 VRHE had no corresponding oxidation peak, which is only observed in CO2-saturated 0.5 M KHCO3 aqueous solution, not in the Ar-saturated 0.5 M KHCO3 aqueous solution (see Supporting Information Figure S16). According to previous studies, this shoulder peak possibly came from CO2 reduction to formic acid.11 Moreover, there was no shoulder reduction peak at −1.07 VRHE on Zn(OH)x but on CuSn (see Figure 2f and Supporting Information Figure S17), which further suggested that the active sites for CO2 reduction are Sn(OH)x rather than Zn(OH)x. The results were also in good agreement with Figure 2b. The shoulder reduction peak at −1.07 VRHE decreased remarkably after removing most of Zn(OH)x on CuSn/Zn(OH)x. However, the CuSn/Zn(OH)x after etching still indicated higher CO2 reduction current than CuSn because there was still a little residual Zn(OH)x in the composite sample (see Supporting Information Table S2). The potential with the maximum FE enhancement of HCOOH on CuSn/Zn(OH)x was −0.7 VRHE, close to −0.81 VRHE of Zn2+/Zn. The potentials of the FE enhancement were in good agreement with the Faradaic potential window of Zn(OH)x. The results suggested that the Faradaic reaction also occurs on Zn(OH)x during CO2 reduction. It is possible that the activity improvement was related to the Faradaic reaction of Zn(OH)x. Moreover, we measured CO2 adsorption of CuSn and CuSn/Zn(OH)x by the BET method, and the results are shown in Supporting Information Figure S18. CuSn/Zn(OH)x indicated higher CO2 adsorption than CuSn, suggesting that the Zn(OH)x can improve the adsorption of CO2 molecules. According to the above results and analysis, we deduce that CO2 molecules are adsorbed and activated on the extrinsic Faradaic layer of Zn(OH)x, and then the intermediates are spilled over onto the active sites of the intrinsic Faradaic layer of Sn(OH)x to produce the final product of HCOOH. In addition to selectivity and activity, stability is also very important for an electrocatalyst. Therefore, the stability of CuSn and CuSn/Zn(OH)x were measured at −0.9 VRHE in CO2-saturated 0.5 M KHCO3 aqueous solution, and the results are shown in Figures 3a–3c. During the stability measurement, the total current density of CuSn/Zn(OH)x increased from 9 to 15 mA/cm2, while the FE for HCOOH kept at about 80% for 40 h. On the CuSn sample, though the total current density also increased from 5 to 9 mA/cm2, the FE for HCOOH decreased from 60% to 40%. Therefore, the extrinsic Faradaic layer of Zn(OH)x improved not only the selectivity and activity, but also the stability of CuSn for CO2 reduction. To investigate the reasons for the improved FE stability, the partial current densities of CuSn and CuSn/Zn(OH)x were measured, and the results are shown in Figure 3c. On the CuSn/Zn(OH)x sample, the partial current density of HCOOH increased from 7 to 11 mA/cm2. In contrast, the partial current density of HCOOH on the CuSn sample remained stable (about 3.5 mA/cm2) for 20 h of measurement. Figure 3d indicates the ratio of Sn/Cu on CuSn and CuSn/Zn(OH)x after 3 and 20 h. The ratio of the Sn/Cu on the CuSn sample increased from 0.57 to 1.01, while the ratio of Sn/Cu on CuSn/Zn(OH)x remained about 0.77 during the stability measurement. From Supporting Information Figure S2, there are Sn0 and Sn (II/IV) on the surface of a CuSn sample, which are assigned to CuSn and Sn(OH)x, respectively. During electrocatalysis, obvious segregation of Sn(OH)x was observed on the surface on the CuSn electrocatalyst, which came from oxidation corrosion of the intermetallic CuSn. Moreover, the higher Cu/Sn ratio on the surface tended to reduce H+ into H2.23,24,34 Therefore, the hydrogen evolution rate increased on CuSn sample during the stability measurement, which decreased the FE for HCOOH (see Supporting Information Figure S19). However, in the CuSn/Zn(OH)x sample, intermetallic CuSn was covered completely by Zn(OH)x (see Supporting Information Figure S3), which prevented the corrosion of the CuSn and the segregation of Sn(OH)x during the stability measurement. Therefore, Zn(OH)x also played a role as a protective layer during CO2 reduction. Figure 3 | (a) Total current density, (b) FE, and (c) partial current density for formic acid on CuSn and CuSn/Zn(OH)x during stability measurement. (d) The ratios of Sn/Cu on the surfaces of CuSn and CuSn/Zn(OH)x after measurement for 3 and 20 h by XPS. SEM images of CuSn after measurement for 3 (e) and 20 h (f), CuSn/Zn(OH)x after 3 (g) and 20 h (h). Download figure Download PowerPoint The morphology stability of CuSn and CuSn/Zn(OH)x were also investigated by SEM after electrocatalytic reaction and the results are shown in Figures 3e–3h. The morphology of CuSn did not change obviously during electrocatalytic reaction, while the surface morphology of CuSn/Zn(OH)x did change obviously with increasing reaction time. Before electrocatalytic reaction, CuSn/Zn(OH)x were spherical particles (see Figure 1b). Dendrite Zn(OH)x was observed on the surface of CuSn/Zn(OH)x after 3 h reaction, which further became to sheet after 20 h measurement. The morphology change of Zn(OH)x during electrocatalytic reaction possibly came from dissolution and redeposition of Zn(OH)x, similar to Cu2+ in a previous study.35 Though obvious morphology changes of Zn(OH)x were observed on CuSn/Zn(OH)x during the stability measurement, the FE for HCOOH did not change, which further suggested that not Zn(OH)x but CuSn provided the active sites for CO2 reduction. Based on the above analysis, a possible mechanism is proposed to understand the roles of Zn(OH)x (see Figure 4). During CO2 reduction, the Sn(OH)x layer on the surface of intermetallic CuSn provides active sites for HCOOH production by a fast and reversible Faradaic reaction, which is considered as an intrinsic Faradaic layer. Zn(OH)x can improve the adsorption and activation of CO2 molecules by a Faradaic reaction, and then spill over the intermediates onto the Sn(OH)x intrinsic Faradaic layer to further reduce into HCOOH, which improves the electrocatalytic activity of CuSn. Therefore, Zn(OH)x can be considered as an extrinsic Faradaic layer. Moreover, Zn(OH)x covers Sn0 on the surface of CuSn completely and prevents the corrosion of the CuSn, suppressing the H2 evolution reaction and improving the selectivity and stability of CO2 reduction remarkably. In the CuSn/Zn(OH)x composite electrocatalyst, an intrinsic Faradaic layer of Sn(OH)x on the surface of CuSn metal makes contact with an extrinsic Faradaic layer of Zn(OH)x. During CO2 reduction, a coupled electron-ion reaction happens at the interface between Sn(OH)x and Zn(OH)x, which spill over the CO2 intermediates from Zn(OH)x to Sn(OH)x. Consequently, CuSn/Zn(OH)x can be considered as a Faradaic junction. In previous studies, only the semiconductor/Faradaic material interface was considered a Faradaic junction.14,16 Since then an intrinsic Faradaic layer has been observed not only on a semiconductor15 but also a metal. A Faradaic junction in fact is an interface between one Faradaic layer and another. Figure 4 | A possible mechanism of Zn(OH)x simultaneously improves the activity, selectivity, and stability of CuSn for electrocatalytic CO2 reduction. Download figure Download PowerPoint Conclusions For the first time, we introduced an extrinsic Faradaic layer Zn(OH)x onto a CuSn electrocatalyst to simultaneously improve the activity, selectivity, and stability of CO2 reduction. CuSn/Zn(OH)x achieves FE of about 90% for HCOOH at the potential of −1.0 VRHE and stable partial current density of 11 mA/cm2 for 40 h, which indicates relatively high performance among Sn-based electrocatalysts (see Supporting Information Table S3). The potentials of the FE enhancement in CuSn/Zn(OH)x are in good agreement with the Faradaic potential window of Zn(OH)x. During CO2 reduction, the intrinsic Faradaic layer of Sn(OH)x on CuSn provides active sites for HCOOH production. In contrast, the extrinsic Faradaic layer of Zn(OH)x plays multiple roles as an adsorption/activation layer, a cover layer, and a protective layer, which can simultaneously improve the activity, selectivity, and stability of CuSn for electrocatalytic CO2 reduction. The new strategy of introducing an extrinsic Faradaic layer on the surface of an electrocatalyst can deepen understanding of the electrocatalytic process and offers guidance to design other high-performance electrocatalysts. Supporting Information Supporting Information is available and includes Figures S1–S19 and Tables S1–S3. Conflicts of Interest There are no conflicts of interest to declare. Author Contributions W.L. supervised the project, proposed the concept, and designed the experiments. F.R. carried out sample preparation, characterization, and electrochemical measurements. W.L. and F.R. analyzed the data and wrote the paper. All the authors discussed the results and gave comments on the manuscript. Acknowledgments The authors thank Dr. Sheng Chu at Southeast University for helpful suggestions. This work was supported by the National Key R&D Program of China (nos. 2017YFE0120700 and 2018YFE0208500), the National Natural Science Foundation of China (nos. 21875105 and 51972164), the National Scientific Instrument Development Major Project of National Natural Science Foundation of China (no. 51627810), and Foshan Xianhu Laboratory of the Advanced Energy Science and Technology Guangdong Laboratory (no. XHD2020-002).