Abstract
Acidic CO2 electroreduction (CO2R) using renewable electricity holds promise for high-efficiency generation of storable liquid chemicals with up to 100% CO2 utilization. However, the strong parasitic hydrogen evolution reaction (HER) limits its selectivity and energy efficiency (EE), especially at ampere-level current densities. Here we present that enhancing CO2R intermediate coverage on catalysts promotes CO2R and concurrently suppresses HER. We identified and engineered robust Cu6Sn5 catalysts with strong *OCHO affinity and weak *H binding, achieving 91% Faradaic efficiency (FE) for formic acid (FA) production at 1.2 A cm−2 and pH 1. Notably, the single-pass carbon efficiency reaches a new benchmark of 77.4% at 0.5 A cm−2 over 300 hours. In situ electrochemical Fourier-transform infrared spectroscopy revealed Cu6Sn5 enhances *OCHO coverage ~2.8× compared to Sn at pH 1. Using a cation-free, solid-state-electrolyte-based membrane-electrode-assembly, we produce 0.36 M pure FA at 88% FE over 130 hours with a marked full-cell EE of 37%.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Introduction
The electrochemical reduction of CO2 (CO2R) to valuable short-chain liquid feedstocks presents an elegant solution for storing intermittent renewable electricity, while also mitigating excessive CO2 emissions resulting from the consumption of fossil fuels1,2,3. In recent decades, significant progress has been made in understanding CO2R reaction pathways on catalysts and the associated energetics to control CO2R selectivity, and in developing gas-diffusion-electrode-based membrane-electrode-assembly (MEA) electrolyzers capable of achieving industrially relevant CO2R current densities over hundreds of mA cm−2 4,5,6,7. Despite these efforts, further advancements in CO2R technology will depend on improving full-cell energy efficiency (i.e., power-to-product efficiency) and maximizing CO2 utilization efficiency (i.e., CO2-to-product conversion).
Operating CO2R in acidic electrolytes offers a potential approach to address significant CO2 losses caused by the reaction between CO2 and OH− in electrolytes to form CO32− during CO2R in alkaline or neutral media. In strong acids with an electrolyte pH ≤ 1, carbonate formation is rare8. However, the kinetics of CO2R under such highly acidic conditions become retarded. Additionally, the substantial increase in surface *H coverage on catalyst surfaces causes a strong competing hydrogen evolution reaction (HER), resulting in a largely reduced Faradaic efficiency (FE) for CO2R9,10.
Early studies have shown that increasing the local concentrations of alkali metal cations in close proximity to catalyst surfaces can enhance CO2R by stabilizing the surface-adsorbed CO2 and CO2R-related intermediates through non-covalent interactions, and, at the same time, impeding the diffusion of protons to the catalyst surfaces11,12,13. However, when the system operates at high potentials (i.e. high current densities), the electroconvective flows near the electrodes become turbulent and disordered. According to the Nernst-Planck equation, this turbulence can cause a fluctuation in cation concentrations near the electrode surface, leading to variations in the local electrical field that may have the potential to disrupt the beneficial cation effect on CO2R14. Prior studies have shown that applying a micrometer-thick layer of nanoparticles as a surface coating layer on catalysts mitigates the irregularities in electroconvective flows within the coated microstructures, which helps maintain a steady concentration profile of alkali metal cations15,16,17. Despite this advancement in mitigating HER, the high local concentration of alkali metal cations can lead to bicarbonate precipitation in the gas diffusion electrodes. The accumulation of bicarbonates hinders CO2 diffusion and hampers CO2R efficiency. To address the bicarbonate precipitation challenge, strategies involve modifying catalysts with immobilized cation groups18 or exploring CO2R in cation-free systems such as using solid-state-electrolyte (SSE) based MEA electrolyzers19. One priority for implementing these approaches is to develop efficient catalysts that can inherently promote CO2R over the competing HER, particularly under strongly acidic conditions.
In this study, we employed computational investigations to discover a Cu6Sn5 catalyst with a notable abundance of surface adsorption sites having a large binding energy difference between *OCHO (the intermediate for formic acid (FA) production) and *COOH (the intermediate for CO production) while also having weaker binding to *H. This configuration allows for maximum selectivity in favor of producing FA over CO and other hydrocarbons while simultaneously suppressing HER.
We experimentally constructed Cu6Sn5 alloy catalysts on polytetrafluoroethylene (PTFE) gas diffusion electrodes using thermal vapor deposition. In situ electrochemical attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy analysis revealed a notable enhancement showcasing a ~2.8 elevation in *OCHO coverage on Cu6Sn5 compared to control catalysts of Sn, operating under identical electrochemical conditions at pH = 1. The active Cu6Sn5 achieves a high FE of 91% for FA production at a current density of 1.2 A cm−2 in a strongly acidic electrolyte at pH = 1. Furthermore, it shows a marked single-pass carbon efficiency (SPCE) of 77.4% at 0.5 A cm−2, representing approximately 2.8-fold enhancement in the SPCE compared to the prior report of ~27.4% obtained at ~0.24 A cm−2 20. Such notable performance was stable over 300 h of continuous acidic CO2R at 0.5 A cm−2 in a flow-cell electrolyzer at pH = 1.
When integrating Cu6Sn5 into a cation-free, SSE-based MEA electrolyzer, we produced 2.6 liters of 0.36 M pure FA at 100 mA cm−2 (area: 4 cm2) over 130 h at a production rate of 5 mL cm−2 h −1, along with a full-cell energy efficiency of 37%. These results feature the potential of enhancing surface intermediate coverage as an efficient means for acidic CO2R, particularly in cation-free systems at industrially relevant current densities.
Results
Density functional theory (DFT) studies
Copper (Cu) has established itself as a predominant and cost-effective electrocatalyst capable of generating a variety of hydrocarbons through CO2R. Previous research has revealed effective strategies for modifying the *CO-binding and *OCHO-binding properties on surfaces of Cu-based alloy surfaces by introducing foreign elements (e.g., Zn, Al, Pb) into the Cu lattice. Specifically, the inclusion of 5–10 at.% Zn or Al into Cu has shown the ability to partially weaken *CO adsorption on the Al or Zn modified Cu site compared to the adjacent Cu-Cu site, thereby creating asymmetric *CO binding energies for improved C2+ production16,17. Additionally, the introduction of a single Pb atom into Cu has been observed to enhance selectivity towards formic acid21. These alternations selectively promote the production of CO/C2+ or FA via distinctive reaction pathways (Fig. 1a). To further curb the competing HER in strong acids, two major approaches are considered: (1) employing alkali metal cations to hamper H+ diffusion toward catalyst surfaces, and (2) augmenting the coverage of surface CO2R intermediates to outcompete the HER.
As a representative p-block metal, Sn has a relatively stronger oxygen affinity and weaker adsorption energy for *H than pristine Cu. It can also form alloys with Cu across the entire concentration profile ranging from 0% to 100%. Interestingly, a previous study has investigated the relationship between CO2R-to-FA activity and the *OCHO adsorption energy with various metal catalysts22, and identified Sn and Cu as promising candidates for selective FA production due to their favorable *OCHO adsorption energies. With this insight, we aimed to create Cu-Sn alloys that could further fine-tune the *OCHO adsorption close to the optimal adsorption-energy values in the adsorption-activity trend. This would contribute to the improvement of the selectivity and activity in the CO2R-to-FA conversion.
To understand the role of Sn in the Cu1−xSnx (x = 0.14, 0.44) catalysts in tuning the catalytic selectivity of CO2R to CO and FA, we built surfaces with various Cu/Sn ratios from pure Cu to Cu1−xSnx alloy, and then to pure Sn, and used DFT to calculate the Gibbs free energetic diagrams for CO and FA generation. Based on the previous literature reviews23,24, the potential rate-determining step (RDS) for the CO2R-to-CO conversion is the first proton transfer to form *COOH from CO2 (CO2 (gas) + H+ + e− + * → *COOH), while the potential RDS for CO2R-to-FA is the first proton transfer to form *OCHO (CO2 (gas) + H+ + e− + * → *OCHO). By using DFT calculations, we identified that the Cu6Sn5 alloy possesses a high abundance of surface sites with a stronger binding affinity to *OCHO compared to *COOH, making it thermodynamically more favorable for producing FA (Fig. 1b). In addition, Cu6Sn5 exhibits notably weaker binding to *H ( + 0.33 eV referring to 1/2 H2), resulting in a substantial energy requirement for H2 production and thus, suppressing HER.
To study how the electronic properties of the surface-active sites on Cu1−xSnx catalysts with varying Sn concentrations affect the selectivity of FA and CO production, we calculated the differential charge density of *OCHO and *COOH over the experimentally observed surfaces in Fig. 1c. Our results indicate that the Cu6Sn5 (−113) surface exhibits more charge transfer with *OCHO than that over other examined surfaces. This charge analysis is consistent with the adsorption energy calculation, where the Cu6Sn5 (−113) surface exhibits the strongest adsorption energy of *OCHO among all the examined surfaces. Such strong adsorption energy of *OCHO also indicates a potentially high surface coverage of *OCHO over the Cu6Sn5 (−113) surface. In addition, since two O−Cu bonds are formed between Oδ− of *OCHO and Cu atoms on Cu1−xSnx surfaces, our results show that Cu6Sn5 (−113) surface presents ~1.3 and ~2.3 eV more favorable formation energy for *OCHO than for *COOH under alkaline (Supplementary Fig. 1) and acidic (Fig. 2a) conditions, respectively. Furthermore, according to the experimental results in Fig. 2b, the production of C2+ and CH4 products disappeared over the metal surface when the concentration of Sn in Cu1−xSnx catalysts was increased to ~14%. To reveal the role of Sn in Cu1−xSnx catalysts in eliminating the C2+ and CH4 production during CO2R, we compared the adsorption energy of CO over pure Cu (111) and Cu0.86Sn0.14 (111) surfaces (Supplementary Fig. 2). The DFT results show that the presence of Sn on the Cu0.86Sn0.14 (111) surface weakens the adsorption of CO to ~0 eV compared to −0.08 eV on the pure Cu (111) surface. Thus, CO will likely desorb from the Cu0.86Sn0.14 (111) surface and be unable to hydrogenate via H protonation to produce CH4 or couple with another adsorbed *CO intermediate to generate C2+ species. These findings are in agreement with previous literature reports and experiments that have observed no C2+ and CH4 species produced over Cu0.86Sn0.1425,26.
We performed Gibbs free energy calculations to quantitatively determine the reaction energy difference between the reaction pathways of CO2–*COOH–CO/C2+ and CO2–*OCHO–HCOO− over the experimentally observed surfaces, including Cu (111), Cu0.86Sn0.14 (111), Cu6Sn5 (−113), and Sn (100) surfaces, where the possible reaction mechanisms have been shown in Supplementary Fig. 3. These calculated surfaces are the dominant facets in XRD measurement in Supplementary Figs. 4–9 and Supplementary Table 1. Gibbs free energetic diagrams for the formation of *OCHO and *COOH over different catalytic surfaces were constructed according to the experimental conditions of room temperature, pH 1 of the electrolyte, and an applied potential of −2.1 V vs. RHE (VRHE). Our theoretical results (Fig. 2a, Supplementary Figs. 10–33) show that the Gibbs free energy differences (ΔG = G(*COOH) − G(*OCHO)) of the potential RDSs of FA and CO pathways during CO2R present a volcano-like plot as increasing the concentration of Sn in the Cu1−xSnx catalysts. Among the four examined surfaces, Cu6Sn5 (−113) shows the largest ΔG between the CO2R to FA and CO pathways, indicating that Cu6Sn5 (−113) contributes the highest selectivity for FA production. The theoretical results are consistent with the experimental results (Fig. 2b, c) that the FE of FA production over Cu6Sn5 is the highest.
To examine the competitive HER over Cu (111), Cu0.86Sn0.14 (111), Cu6Sn5 (−113) and Sn (100) surfaces, we examined the energy diagram of HER (Supplementary Fig. 34). The corresponding *H adsorption configurations can be seen in Supplementary Figs. 35–38. The increased free energy of HER indicates that the HER will be suppressed, and the selectivity for CO2R can thus be improved over Cu0.86Sn0.14, Cu6Sn5, and Sn compared to that on pristine Cu. This is consistent with experimental results (Fig. 2b), which show the FE of HER over Cu is higher than that on Cu0.86Sn0.14, Cu6Sn5, and Sn.
Finally, we built surfaces with different Cu/Sn ratios, ranging from pure Cu to Cu1−xSnx alloy, and then to pure Sn. Using DFT calculations, we analyzed the surface formation energetics for each surface. As depicted in Supplementary Fig. 39, Cu6Sn5 (−113) shows improved stability with a lower surface formation energy compared to the pristine Cu (111) facet. Conversely, Sn (100) shows poor stability with a higher surface formation energy.
Electrochemical CO2R studies
With the theoretical insight, we conducted a systematical investigation into the Cu–Sn alloy as an illustrative catalyst system with the objective of obtaining exclusive CO2R to HCOOH at high current densities while also achieving high energy and carbon efficiencies. Experimentally, we created Cu1−xSnx (x = 0.08, 0.14, 0.28, 0.44, 0.71, 0.88) and control catalysts of pure Cu and Sn on PTFE gas diffusion electrodes using thermal evaporation (Supplementary Figs. 4 and 40, 41). We evaluated their CO2R performance for Cu1−xSnx (x = 0.08, 0.14, 0.28, 0.44, 0.71, 0.88) and controlled Cu and Sn catalysts in a flow cell with a three-electrode configuration in both alkaline and acid electrolytes. We quantitatively analyzed the gas and liquid products using gas chromatography (GC), nuclear magnetic resonance spectroscopy (NMR), and ion chromatography (IC).
In a 1 M KOH electrolyte at pH 14, the linear sweep voltammetry (LSV) curves in Supplementary Fig. 42 showed a sharp increase in the current densities with the increase of applied negative potentials, indicating efficient electrical conductivity for the Cu and Cu1−xSnx (x < 0.71) electrodes. With the addition of Cu in Cu1−xSnx (x < 0.71), the onset potential for formate shifted positively, indicating improved CO2R kinetics. At high current densities, Cu6Sn5 also displayed the highest selectivity for formate production, with over 90% formate FEs across a wide current density range of 0.4 to 1.2 A cm−2 (Supplementary Fig. 43). The Cu6Sn5 catalyst showed formate partial current densities of up to ~1.5 A cm−2 at a relatively low operating potential of around −1.8 VRHE (Supplementary Fig. 44) and the highest formate selectivity close to 90% (Supplementary Fig. 43).
To assess the stability of Cu6Sn5 in 1 M KOH, we carried out Galvanostatic tests using an alternating current density mode (0.05 A cm−2 for 30 s and 0.5 A cm−2 for 90 s)17 for 160 hours in a flow cell (Supplementary Fig. 45). Prior to the stability test, we coated a carbon-Nafion™ or SiC-Nafion™ mixed layer with a thickness of 2–5 μm on the surface of Cu6Sn5 to enable a uniform electrohydrodynamic flow near the catalysts14. Through the 120-h CO2R operation at a current density of 0.5 A cm−2, we obtained a stable formate FE of above 85%. When comparing the FE and current density of this work with the previous reports under alkaline conditions, we found that we found that our results are superior (Supplementary Fig. 46 and Supplementary Table 2)6,11,19,21,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44.
In 3 M KCl and 0.05 M H2SO4 electrolyte at pH 1, Cu6Sn5 exhibited the highest selectivity, reaching up to 96% at −1.4 VRHE for formic acid in Fig. 2c. Remarkably, Cu6Sn5 showed a high FE of above 90% for FA production across a wide range of current densities from 0.4 to 1.2 A cm−2 (Fig. 2b). We examined the SPCE for Cu6Sn5 under a constant current density of 0.5 A cm−2 in a flow cell equipped with a 1.7 × 1.7 cm2 serpentine channel reaction area at various CO2 flow rates and electrolyte pH levels (more details are presented in the Supplementary Information)15. Figure 2d presents a maximum SPCE of 77.4%, achieved at a CO2 flow rate of 3 standard cubic centimeter per minute (sccm) under pH 1 conditions. The electrode of Cu6Sn5 coated with carbon-Nafion™ or SiC-Nafion™ demonstrated stable FA production for over 300 hours, maintaining over 70% FE at a cathodic potential of −2.5 VRHE. This equated to a 40% cathodic energy efficiency (CEE) at 0.5 A cm−2 at pH 1 (Supplementary Fig. 47). Theoretical investigations were correspondingly carried out in this segment to investigate the effects of the electrochemical operating environment effects (e.g., electrolyte of 3 M K+, applied potential of −2.1 VRHE, and pH = 1) on the selectivity for FA vs. CO formation using Constant Electrode Potential (CEP) model via performing Grand canonical DFT calculations45,46,47,48,49. The selectivity for FA across Cu (111), Cu0.86Sn0.14 (111), Cu6Sn5 (−113), and Sn (100) surfaces exhibits a volcano plot (Fig. 2a and Supplementary Fig. 48), with Cu6Sn5 (−113) displaying the highest energy difference between the two pivotal intermediates, favoring the adsorption of *OCHO over CO. More details regarding grand canonical DFT simulations are provided in Supplementary Information. We compared this work with previously reported results on the CO2R-to-FA production under acidic conditions in terms of elevated current density, FE, SPCE, and stability in Fig. 2e12,15,20.
Characterization
We conducted material and structure characterizations, as well as surface wettability analysis, for the Cu6Sn5 catalyst before the CO2R reaction. Figure 3a shows a schematic of the CO2R process on Cu6Sn5/PTFE electrodes. The cross-sectional scanning electron microscope (SEM) image in Fig. 3b shows the well-defined structure of Cu6Sn5 nano-to-micro particles densely packed on the PTFE substrate. This Cu6Sn5-on-PTFE structure was synthesized over a large scale via thermal evaporation (Fig. 3c and Supplementary Fig. 49), exhibiting hydrophobicity, as indicated by a water contact angle of 128° (Fig. 3c). This hydrophobicity enables CO2 diffusion through the gaps between the particles on the PTFE sides to the catalyst surface (Fig. 3a). The top-view SEM presents that the diameter of the Cu6Sn5 particles is 300–600 nm (Fig. 3d). Energy dispersive X-ray spectroscopy in transmission electron microscopy (STEM-EDX) analysis revealed a uniform distribution of Sn and Cu over the majority of the Cu6Sn5 particles (Fig. 3e).
Detailed structural analysis using transmission electron microscope (TEM), high-resolution transmission electron microscope (HRTEM), and selective area electron diffraction (SAED) analyses (Fig. 3f, g) confirmed the formation of monoclinic Cu6Sn5 crystal. These results are in line with the X-ray powder diffraction (XRD) patterns in Supplementary Fig. 50, which indicate the presence of the main (−113), (132), and (−314) facets for Cu6Sn5. More details regarding the crystallographic and material analysis of catalysts after acidic CO2R operation are presented in Supplementary Figs. 51–53.
In situ ATR-FTIR analysis
We used in situ ATR-FTIR spectroscopy to study CO2R intermediates coverage on Cu6Sn5-on-PTFE, Cu-on-PTFE, and Sn-on-PTFE electrodes during acidic CO2R electrolysis in the same pH 1 electrolyte of 3 M KCl and 0.05 M H2SO4, spanning a range of applied potentials from −0.24 to −1.64 VRHE. We also incorporated an internal standard of potassium ferricyanide in the electrolytes to facilitate peak area calibration. In the ATR-FTIR spectra for the Cu6Sn5 and Sn catalysts (Fig. 4a–c), significant peaks were absent within the range of 1900 to 2200 cm−1. In contrast, distinct infrared vibration bands related to the *OCHO intermediate were observed within the 1375–1385 cm−1 range for all Cu6Sn5, Cu, and Sn catalysts. The onset potential for the *OCHO intermediate on Cu6Sn5 was −0.44 VRHE, lower than that on Cu and Sn, indicating fast CO2R-to-HCOOH kinetics with Cu6Sn5. During the in situ ATR-FTIR test using a 5 mM potassium ferricyanide internal standard in 3 M KCl and 0.05 M H2SO4, we observed a constant peak area for potassium ferricyanide within the potential range of −0.54 to −0.94 VRHE. Consequently, we established a calibration curve within this potential range, using the potassium ferricyanide peak as a reference. This allowed us to calculate the peak areas of *OCHO intermediates on different catalysts at the same potential range (Fig. 4d, e and Supplementary Table 3). Notably, adhering the established calibration relationship, the plateau *OCHO peak intensity on Cu6Sn5 measured at potentials above −1.04 VRHE in the CO2R electrolyte is ~2.8 times higher than that of Sn, with all measurements conducted under identical conditions (Fig. 4f). These results point toward an enhanced surface coverage of *OCHO on Cu6Sn5, which facilitates selective FA production during acidic CO2R.
To produce a pure formic acid solution, we conducted CO2R using an AEM (Sustainion X37-50 Grade) based SSE-MEA electrolyzer. 0.5 M H2SO4 was used as the anolyte and IrOx/Ti foam was used as the anode (Fig. 5a). During the electrolysis, HCOO− was generated on the cathode side and diffused through the AEM into the middle SSE layer. Simultaneously, H+ permeated through the CEM (Nafion 117) and reacted with HCOO− to form FA. The produced FA solution was collected by passing deionized water through the SSE layer. As shown in Fig. 5a, the operating potential is 3.4 V for the overall CO2-to-FA and H2O-to-O2 reaction in an SSE-based MEA electrolyzer. The detailed potential distributions are listed in Fig. 5a: a theoretical potential of 1.43 V required to initiate the CO2R-to-FA reaction, a ~0.3 V overpotential on the IrOx loaded Ti mesh anode, and an ~1.5 V overpotential on the SSE, anion and cation membranes, and Cu6Sn5 cathode. In Fig. 5b, using the Cu6Sn5 catalyst, the FA FE reached ~96% at 100 mA cm−2. The full-cell voltage was −3.7 V and the full-cell energy efficiency was over 37%. We evaluated the CO2R performance in different anolytes (0.1 M, 0.5 M, and 1 M H2SO4) in Supplementary Fig. 54. We quantified the amount of FA produced using IC, NMR, and pH measurements (Supplementary Fig. 55). The results confirmed the production of 2.6 liters of 0.36 M pure FA solution (Supplementary Fig. 56) with a production rate of 5 mL cm−2 h−1 (20 mL h−1) over a continuous 130-h CO2R process (Fig. 5c). Figure 5d shows an optical image of an SSE-based MEA and a steady production rate of FA over the 130-h CO2R process to produce ~700 mmol pure FA.
Discussion
In this study, we demonstrate the effectiveness of enhancing surface coverage of CO2R intermediates as a strategy to promote CO2R at ampere-level current densities and under strongly acidic conditions. Theoretical DFT calculations have revealed that Cu6Sn5 enhances the adsorption of *OCHO compared to *COOH and *H, resulting in increased surface *OCHO coverage and promoted FA production in CO2R while simultaneously suppressing the HER. Experiments have identified a robust Cu6Sn5 catalyst that exhibits exclusive FA production with over 90% FE at 1.2 A cm−2, and a remarkable 77.4% carbon efficiency at 0.5 A cm−2, maintaining stability over 300 h of continuous CO2R operation at pH = 1. In situ electrochemical ATR-FTIR spectroscopy validates an approximately 2.8× enhancement in *OCHO coverage on Cu6Sn5 compared to the control catalyst of Sn, operating under identical electrochemical conditions at pH = 1. Furthermore, using an SSE-based MEA electrolyzer, we stably produce pure FA solution with a concentration of 0.36 M over a 130-hour reaction at a full-cell energy efficiency surpassing 37%. We expect that the insights gained from our work, particularly in the context of optimizing intermediate adsorption and coverage, will provide valuable guidance for the advancement of selective and energy-efficient CO2R with long-lasting performance in cation-free, MEA-based electrolyzers.
Methods
Density functional theory (DFT) calculation
DFT calculations were conducted using the Vienna Ab-initio Simulation Package (VASP)50. To take the solvation effects into consideration, hybrid solvation effects have been employed in VASPsol along with explicit water molecules at the interface45,46,47,48. The Constant Electrode Potential (CEP) model has been employed to investigate the constant negative applied potential and cation effects through grand-canonical DFT (GC-DFT) calculations49. These calculations were used to compare the energy diagrams of two possible CO2 reduction reaction (CO2R) pathways over different surfaces. We have investigated Cu (111), Cu0.86Sn0.14 (111), Cu6Sn5 (−113), and Sn (100) surfaces for calculations (Supplementary Fig. 4), where these facets are also the most thermodynamic favorable ones for Cu, Sn, and Cu1−xSnx alloy systems. More DFT calculation details are included in the Supplementary Information.
Synthesis
We fabricated a series of Cu1−xSnx (x = 0.08, 0.14, 0.28, 0.44, 0.71, 0.88), Cu, and Sn electrocatalysts by thermal evaporation (SKY-RH400). Cu and Sn particles were placed separately in two Molybdenum boats inside the deposition chamber and melted slowly under the pressure of 10−5 Torr for thermal evaporation. We controlled the thermal evaporation rates of Cu and Sn to adjust the Cu/Sn ratios in the synthesized alloys. Evaporating rates of approximately x Å s−1 (x = 0.08, 0.14, 0.28, 0.44, 0.71, 0.88) for Sn and approximately (1−x) Å s−1 for Cu were used to produce alloyed Cu1−xSnx films on the polytetrafluoroethylene (PTFE) substrates. A quartz crystal monitor was used to observe the thicknesses of evaporated Cu1−xSnx. Evaporation of pure Sn and pure Cu electrocatalysts was conducted using a similar fabrication procedure. The loading of all catalysts is 800 nm thick.
Characterization
Transmission electron microscopy (SEM) images were taken using a Hitachi SU8100 SEM at an accelerating voltage of 5 kV. High-resolution transmission electron microscopy (HRTEM) and transmission electron microscopy-energy dispersive X-ray spectroscopy (TEM-EDX), selected area electron diffraction (SAED), and bright-field and dark-field TEM analyses were performed in a TEM (Tecn F20) with an accelerating voltage of 200 kV. X-ray powder diffraction (XRD) was carried out with a Bruker D8 Advance at a scanning rate of 10° min−1 in the 2θ range from 20° to 80°. X-ray photoelectron spectroscopy (XPS) studies were performed using PHI5000 VersaProbe. The binding energy data were calibrated relative to the C 1 s signal at 284.6 eV.
Electrochemical experiments
Experiments under alkaline and acidic conditions were performed in a flow cell using a three-electrode system. Ag/AgCl electrode was used as the reference electrode, commercial Ni foam (for use under alkaline conditions) and Pt (for use under acidic conditions) were used as the counter electrodes, Cu1−xSnx (x = 0.08, 0.14, 0.28, 0.44, 0.71, 0.88), Cu, and Sn on PTFE electrodes were used as the working electrodes (area: 0.5 cm2), and the electrolytes were 1 M KOH (for alkaline CO2 electrolysis) and 0.05 M H2SO4 with 3 M KCl (for acidic CO2 electrolysis). Anion exchange membrane (Fumasep FAB-PK-130, size: 2 × 2 cm2, thickness: 130 μm) and proton exchange membrane (Nafion N117, size: 2 × 2 cm2, thickness: 183 μm) were used as the ion exchange membranes. The proton exchange membrane was immersed in 0.5 M H2SO4 for about 2 h before use. Experiments using solid-state electrolyte (SSE) were performed in a membrane electrode assembly (MEA) system with a 1 mm-thick SSE layer between the cathode and anode. IrOx/Ti foam was used as the anode, and Cu6Sn5 was used as the cathode (area: 4 cm2). Anion exchange membrane (Sustainion X37-50 Grade, size: 4 × 4 cm2, thickness: 50 μm) and proton exchange membrane (Nafion N117, size: 4 × 4 cm2, thickness: 183 μm) were used as the ion exchange membranes. The SSE was AmberChrom 50WX4 hydrogen form (J&K Scientific), and the anolyte was 0.5 M H2SO4. The electrochemical workstation was Autolab PGSTAT302N (Metrohm). The gas flow rate was controlled at 25 mL min−1 by an electronic flow meter, and the end flow rate was calibrated by a soap film flow meter during the test.
All of the electrode potentials vs. the Ag/AgCl electrode were converted to the potentials vs. reversible hydrogen electrode (RHE) using the following Eq. (1):
Gas-phase products were measured using gas chromatography (GC Agilent 990, Perkin Elmer Clarus 680). According to the peak area, the Faradaic efficiency of the gas products can be obtained, and the calculation formula is as the following Eq. (2):
where F is the Faraday’s constant, which is 96485 C mol−1. z is the number of electrons required to reduce CO2 to a CO or H2 molecule, which is 2. v is the gas flow rate, here is 25 mL min−1. n is the concentration of the gas products obtained by GC with 1 mL of sample gas, the unit is mol mL−1. I is the current applied to the sample, the unit is A. t is the reaction time, and the unit is s.
Liquid product HCOO− was measured using ion chromatography (IC, SH-AC-11, Qingdao shenghan). The FE towards formate or FA at each current density was calculated by adding up both anodic and cathodic FEs. We first obtained a standard curve with a concentration gradient of sodium formate (HCOONa), HCOO− concentrations were set as 1 ppm, 2 ppm, 5 ppm, 10 ppm, 50 ppm, and 100 ppm (mg L−1). According to this, the HCOO− concentration in the electrolyte after the reaction can be obtained. The formula of faradaic efficiency calculation for HCOO− is as the following Eq. (3):
where F is the Faraday’s constant, which is 96485 C mol−1. n is the concentration of HCOO− measured by the instrument based on the standard curve, and the unit is mg L−1. z is the number of electrons required for the reduction of CO2 into HCOO−, here is 2. V is the volume of the catholyte, here is 0.03 L. t is the reaction time, and the unit is s.
All data, including Faradaic efficiencies, were collected based on 1-hour electrolysis. Stability tests were conducted over 300 h. The overpotential is determined by subtracting the operating potential at specific current densities for formic acid generation from the theoretic potential for formic acid production, which is −0.199 VRHE. In situ electrochemical attenuated total reflection Fourier-transform infrared spectroscopy (ATR-FTIR) experiments were conducted on a Thermo Scientific Nicolet 6700 FTIR spectrometer with ZnSe as the prismatic window at room temperature. A three-electrode electrochemical single-cell was used for the tests (Supplementary Fig. 57). The thermally-evaporated Cu6Sn5 catalyst on a carbon gas diffusion layer (Freudenberg H15C13) was used as the working electrode, a Pt wire was used as the counter electrode, and a saturated Ag/AgCl electrode was used as the reference electrode. 3 M KCl and 0.05 M H2SO4 (pH = 1) saturated with CO2 was used as the electrolyte. Open circuit potential (OCP) was conducted as a comparison, and the data was collected using chronoamperometric tests from −0.24 to −1.64 VRHE. The peak area of *OCHO in FTIR was calculated by integrating the corresponding curve areas in the same interval (1250–1500 cm−1) obtained at different potentials. More experimental details are discussed in the Supplementary Information.
Data availability
Source data to generate figures and tables are available from the corresponding authors.
References
Montoya, J. H. et al. Materials for solar fuels and chemicals. Nat. Mater. 16, 70–81 (2017).
Urbain, F. et al. Multijunction Si photocathodes with tunable photovoltages from 2.0 V to 2.8 V for light induced water splitting. Energy Environ. Sci. 9, 145–154 (2016).
Mistry, H. et al. Nanostructured electrocatalysts with tunable activity and selectivity. Nat. Rev. Mater. 1, 16009 (2016).
Ulas, B. & Kivrak, H. D. Direct formic acid and formate fuel cells (DF(A)FCs). in Direct Liquid Fuel Cells. 149–176 (Elsevier, Amsterdam, The Netherlands, 2021).
Pérez-Fortes, M. et al. Formic acid synthesis using CO2 as raw material: Techno-economic and environmental evaluation and market potential. Int. J. Hydrog. Energy 41, 16444–16462 (2016).
Ma, W. et al. Promoting electrocatalytic CO2 reduction to formate via sulfur-boosting water activation on indium surfaces. Nat. Commun. 10, 892 (2019).
Pan, B. et al. Close to 90% single-pass conversion efficiency for CO2 electroreduction in an acid-fed membrane electrode assembly. ACS Energy Lett. 7, 4224–4231 (2022).
Huang, J. E. et al. CO2 electrolysis to multicarbon products in strong acid. Science 372, 1074–1078 (2021).
Jiang, M. et al. Rapid and green electric-explosion preparation of spherical indium nanocrystals with abundant metal defects for highly-selective CO2 electroreduction. Nano Lett. 23, 291–297 (2023).
Yang, Z. et al. P-block metal-based (Sn, In, Bi, Pb) electrocatalysts for selective reduction of CO2 to formate. APL Mater. 8, 060901 (2020).
Yan, S. et al. Electron localization and lattice strain induced by surface lithium doping enable ampere‐level electrosynthesis of formate from CO2. Angew. Chem. Int. Ed. 60, 25741–25745 (2021).
Li, L. et al. Achieving high single‐pass carbon conversion efficiencies in durable CO2 electroreduction in strong acids via electrode structure engineering. Angew. Chem. Int. Ed. 62, e202300226 (2023).
Dinh, C. T. et al. CO2 electroreduction to ethylene via hydroxide-mediated copper catalysis at an abrupt interface. Science 360, 783–787 (2018).
Yang, K. D. et al. Morphology‐directed selective production of ethylene or ethane from CO2 on a Cu mesopore electrode. Angew. Chem. Int. Ed. 129, 814–818 (2017).
Shen, H. et al. Acidic CO2-to-HCOOH electrolysis with industrial-level current on phase engineered tin sulfide. Nat. Commun. 14, 2843 (2023).
Zhong, M. et al. Accelerated discovery of CO2 electrocatalysts using active machine learning. Nature 581, 178–183 (2020).
Zhang, J. et al. Accelerating electrochemical CO2 reduction to multi-carbon products via asymmetric intermediate binding at confined nanointerfaces. Nat. Commun. 14, 1298 (2023).
Fan, M. et al. Cationic-group-functionalized electrocatalysts enable stable acidic CO2 electrolysis. Nat. Catal. 6, 763–772 (2023).
Xia, C. et al. Continuous production of pure liquid fuel solutions via electrocatalytic CO2 reduction using solid-electrolyte devices. Nat. Energy 4, 776–785 (2019).
Qiao, Y. et al. Engineering the local microenvironment over Bi nanosheets for highly selective electrocatalytic conversion of CO2 to HCOOH in strong acid. ACS Catal. 12, 2357–2364 (2022).
Zheng, T. et al. Copper-catalysed exclusive CO2 to pure formic acid conversion via single-atom alloying. Nat. Nanotechnol. 16, 1386–1393 (2021).
Feaster, J. T. et al. Understanding selectivity for the electrochemical reduction of carbon dioxide to formic acid and carbon monoxide on metal electrodes. ACS Catal. 7, 4822–4827 (2017).
Ding, P. et al. Metal-based electrocatalytic conversion of CO2 to formic acid/formate. J. Mater. Chem. A 8, 21947–21960 (2020).
Ren, M. et al. CO2 to formic acid using Cu-Sn on laser-induced graphene. ACS Appl. Mater. Interfaces 12, 41223–41229 (2020).
Morimoto, M. et al. Electrodeposited Cu-Sn alloy for electrochemical CO2 reduction to CO/HCOO−. Electrocatalysis 9, 323–332 (2018).
Li, D. et al. Facile synthesis of porous Cu-Sn alloy electrode with prior selectivity of formate in a wide potential range for CO2 electrochemical reduction. Appl. Catal. B: Environ. 292, 120119 (2021).
Chen, Z. et al. 3D hierarchical porous structured carbon nanotube aerogel-supported Sn spheroidal particles: an efficient and selective catalyst for electrochemical reduction of CO2 to formate. J. Mater. Chem. A 5, 24651–24656 (2017).
Liang, C. et al. High efficiency electrochemical reduction of CO2 beyond the two-electron transfer pathway on grain boundary rich ultra-small SnO2 nanoparticles. J. Mater. Chem. A 6, 10313–10319 (2018).
Fan, L. et al. Electrochemical CO2 reduction to high-concentration pure formic acid solutions in an all-solid-state reactor. Nat. Commun. 11, 3633 (2020).
Grigioni, I. et al. CO2 Electroreduction to formate at a partial current density of 930 mA cm−2 with InP colloidal quantum dot derived catalysts. ACS Energy Lett. 6, 79–84 (2021).
Lei, F. et al. Metallic tin quantum sheets confined in graphene toward high-efficiency carbon dioxide electroreduction. Nat. Commun. 7, 12697 (2016).
García De Arquer, F. P. et al. 2D metal oxyhalide-derived catalysts for efficient CO2 electroreduction. Adv. Mater. 30, 1802858 (2018).
Chen, Y. et al. A robust, scalable platform for the electrochemical conversion of CO2 to formate: identifying pathways to higher energy efficiencies. ACS Energy Lett. 5, 1825–1833 (2020).
Kopljar, D. et al. Transferring electrochemical CO2 reduction from semi-batch into continuous operation mode using gas diffusion electrodes. Chem. Eng. Technol. 39, 2042–2050 (2016).
Löwe, A. et al. Optimizing reaction conditions and gas diffusion electrodes applied in the CO2 reduction reaction to formate to reach current densities up to 1.8 A cm−2. ACS Sustain. Chem. Eng. 9, 4213–4223 (2021).
Del Castillo, A. et al. Sn nanoparticles on gas diffusion electrodes: Synthesis, characterization and use for continuous CO2 electroreduction to formate. J. CO2 Util. 18, 222–228 (2017).
Sen, S. et al. Electroreduction of carbon dioxide to formate at high current densities using tin and tin oxide gas diffusion electrodes. J. Appl. Electrochem. 49, 917–928 (2019).
Wu, D. et al. Boosting formate production at high current density from CO2 electroreduction on defect-rich hierarchical mesoporous Bi/Bi2O3 junction nanosheets. Appl. Catal. B: Environ. 271, 118957 (2020).
Rabiee, H. et al. Tuning the product selectivity of the Cu hollow fiber gas diffusion electrode for efficient CO2 reduction to formate by controlled surface Sn electrodeposition. ACS Appl. Mater. Interfaces 12, 21670–21681 (2020).
Xing, Y. et al. Bi@Sn core–shell structure with compressive strain boosts the electroreduction of CO2 into formic acid. Adv. Sci. 7, 1902989 (2020).
Creissen, C. E. et al. Molecular inhibition for selective CO2 conversion. Angew. Chem. Int. Ed. 61, e202206279 (2022).
Wang, W. et al. In situ phase separation into coupled interfaces for promoting CO2 electroreduction to formate over a wide potential window. Angew. Chem. Int. Ed. 60, 22940–22947 (2021).
Yang, Q. et al. Novel Bi‐doped amorphous SnOx nanoshells for efficient electrochemical CO2 reduction into formate at low overpotentials. Adv. Mater. 32, 2002822 (2020).
Ma, X. et al. Facet dopant regulation of Cu2O boosts electrocatalytic CO2 reduction to formate. Adv. Funct. Mater. 33, 2213145 (2023).
Mathew, K. et al. Implicit self-consistent electrolyte model in plane-wave density-functional theory. J. Chem. Phys. 151, 234101 (2019).
Petrosyan, S. A. et al. Joint density-functional theory: Ab initio study of Cr2O3 surface chemistry in solution. J. Phys. Chem. B 109, 15436–15444 (2005).
Mathew, K. et al. Implicit solvation model for density-functional study of nanocrystal surfaces and reaction pathways. Chem. Phys. 140, 084106 (2014).
Goodpaster, J. D. et al. Identification of possible pathways for C–C bond formation during electrochemical reduction of CO2: new theoretical insights from an improved electrochemical model. J. Phys. Chem. Lett. 7, 1471–1477 (2016).
Hormann, N. G. et al. Grand canonical simulations of electrochemical interfaces in implicit solvation models. J. Chem. Phys. 150, 041730 (2019).
Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B: Condens. Matter 47, 558–561 (1993).
Acknowledgements
The authors thank the National Natural Science Foundation of China (grant number 22272078), the National Key R&D Program of China (No. 2020YFA0406102), the Frontiers Science Center for Critical Earth Material Cycling of Nanjing University, the “Innovation and Entrepreneurship of Talents plan” of Jiangsu Province, and Program for Innovative Talents and Entrepreneurs in Jiangsu (JSSCTD202138). Y.X. and F.C. thank the sponsorship from the Department of Navy under the awards N00014-22-1-2001 and N00014-20-1-2858 issued by the Office of Naval Research. The United States Government has a royalty-free license throughout the world for all copyrightable material contained herein. Y.X. and F.C. acknowledge the computational resources provided by the Massachusetts Green High-Performance Computing Center (MGHPCC), the computational resources provided by ACCESS Maximize Project No. CHM220016, ACCESS Accelerate Project No. CHE200083, and ACCESS Explore Project No. CHE220075.
Author information
Authors and Affiliations
Contributions
M.Z. conceived the idea and designed the experiments. X.Y., L.L., Ming.Z., and W.Q. conducted the synthesis, characterizations, and flow-cell tests. X.Y., L.L., and W.Q. designed and carried out the MEA tests. Y.X. and F.C. carried out DFT calculations and analysis. Y.X. and F.C. wrote the theoretical section of the manuscript. X.Y., L.L., Ming.Z., and M.Z. wrote the experimental and other sections of the manuscript. All authors discussed the results and contributed to the preparation of the manuscript. M.Z. and F.C. supervised the project.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Yu, X., Xu, Y., Li, L. et al. Coverage enhancement accelerates acidic CO2 electrolysis at ampere-level current with high energy and carbon efficiencies. Nat Commun 15, 1711 (2024). https://doi.org/10.1038/s41467-024-45988-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-024-45988-4
- Springer Nature Limited
This article is cited by
-
An acid-tolerant metal-organic framework for industrial CO2 electrolysis using a proton exchange membrane
Nature Communications (2024)
-
Tailoring microenvironment for efficient CO2 electroreduction through nanoconfinement strategy
Nano Research (2024)