Electroreduction of diluted CO₂ to multicarbon products with high carbon utilization at 800 mA cm^–2 in strongly acidic media

Abstract

Acidic CO₂ electroreduction using diluted CO₂ (as in flue gas) as the feedstock can simultaneously circumvent the CO₂ purification step and lower the carbon loss in conventional alkaline or neutral electrolyte, and thus is highly desired but has rarely been achieved thus far. Herein, we report a simple and general strategy using an imidazolium-based anion-exchange ionomer as the coating layer, which could enrich the diluted CO₂ to generate a high local CO₂ concentration, and simultaneously block the proton transport to the cathode surface to suppress the competing hydrogen evolution reaction. As a result, the ionomer-modified Cu catalyst can achieve an efficient electroreduction of diluted CO₂ (15 vol% CO₂) to multicarbon (C₂₊) products in strong acid (pH 0.8), with a high C₂₊ Faradaic efficiency of 70.5% and a high single-pass carbon efficiency of 73.6% at a current density of 800 mA cm^–2, competitive with that obtained with pure CO₂. These findings provide opportunity for the direct electrochemical conversion of flue gas into valuable products with high efficiency.

Introduction

Electrocatalytic carbon dioxide reduction (CO₂R) is a sustainable way to convert CO₂ into value-added chemicals and fuels by using renewable electricity, contributing to the development of a carbon-neutral economy^1,2,3,4,5,6. Currently, most of the related studies have been focused on the use of high-purity CO₂ gas (> 99%) as the feedstock for electrocatalytic CO₂R, whereas industrial flue gas as the primary form of CO₂ emission, contains low-concentration CO₂ gas (~ 15 vol %)^7,8,9,10. This means that we need to capture and enrich the diluted CO₂ to generate high-purity CO₂ for subsequent CO₂R electrolysis¹⁰. However, CO₂ purification from flue gas is an energy-intensive process and usually costs $70–100 per ton of CO₂^8,11,12,13. It would be very promising to directly utilize the diluted CO₂ streams as the feedstock for the cost-effective CO₂ electroreduction into valuable products^7,8,9,10. Unfortunately, when diluted CO₂ gas is used as the feedstock, the competing hydrogen evolution reaction (HER) usually becomes the dominated reaction owing to the low CO₂ concentration, thus resulting in the unsatisfactory CO₂R performance^7,10,14. To date, the direct electroreduction of diluted CO₂ into valuable products with high selectivity still remains a formidable challenge.

Among the diverse CO₂R products, multicarbon (C₂₊) products, including ethylene, ethanol, and acetate, have received broad attention for their high market price and energy density^{3,15,16,17,18}. Up to now, most reported catalysts and reactors for CO₂R-to-C₂₊ conversion mainly operate in alkaline or neutral electrolytes, in which the local alkaline microenvironment can help to activate CO₂ molecules and facilitate the C–C coupling step while suppress the HER^18,19,20,21. However, the input CO₂ suffers from serious consumption by reacting with hydroxide ions to form unnecessary carbonate or bicarbonate ions, leading to a very low single-pass carbon efficiency (SPCE, typically ≤ 25%) for C₂₊ products^22,23. The further regeneration of lost CO₂ from (bi)carbonate requires an additional ~ 50% of input energy²². Using acidic electrolytes provides an alternative route to alleviate the (bi)carbonate formation and carbon loss problem, but the fast reaction rate for kinetically favorable HER usually leads to a low CO₂R Faradaic efficiency (FE) in this condition^24,25,26,27. Recently, enormous efforts and attempts have been devoted to exploring diverse strategies to suppress the HER kinetics for effective CO₂R-to-C₂₊ conversion in acid^{21,22,28,29,30,31,32,33}. However, high-purity CO₂ was still used as the fed gas to improve the selectivity for electrocatalytic CO₂R. The direct utilization of diluted CO₂ (as in flue gas) instead of pure CO₂ to generate C₂₊ products with high FE in strong acid is highly desired but has never been achieved.

For a bare Cu catalyst under the diluted CO₂ condition, its low local CO₂ concentration would lead to the easy occurrence of the kinetically favored HER over CO₂R (Fig. 1a). We postulate that a catalyst containing a proper CO₂ enriching layer can concentrate the diluted CO₂ to generate a high local CO₂ concentration at the catalyst surface. We thus engineered Cu surface with an imidazolium-based anion-exchange ionomer coating layer, which can capture and enrich diluted CO₂ to obtain a high local CO₂ concentration, and simultaneously obstruct the proton transport to suppress HER, thereby forming rich gas–liquid–solid three-phase interfaces for efficient electroreduction of diluted CO₂ in acidic media (Fig. 1a). With diluted CO₂ (15 vol% CO₂) as the feedstock and strongly acidic solution (pH 0.8) as the electrolyte, our catalyst achieved a high C₂₊ FE of 70.5% and a high SPCE of 73.6% at a current density of 800 mA cm^–2, surpassing the reported CO₂R-to-C₂₊ performance achieved in pure CO₂ atmosphere.

Fig. 1: CO₂ capture and transport.

a Schematic illustration of the catalyst with altered microenvironment after coating Sustainion layer for efficient electroreduction of diluted CO₂ in strong acid. b Digital photograph for bare glassy carbon electrode (GCE) in pure water after bubbling CO₂ gas. c, d Digital photographs for GCE@Sustainion in pure water before (c) and after (d) bubbling CO₂ gas. e CO₂ gas bubble adhesive force measurements for GCE and GCE@Sustainion. f, g Real-time CO₂ bubble transport behaviors for GCE (f) and GCE@Sustainion (g).

Results

CO₂ capture and transport

We selected Sustainion ionomer (Fig. 1a), a commercially available imidazolium-based anion-exchange ionomer, to enrich the local CO₂ concentration and increase the CO₂-to-water ratio, on account of the strong CO₂ affinity with imidazolium group^34,35,36,37. A flat glassy carbon electrode (GCE) was used as the substrate for the observation of the CO₂ adsorption behavior of Sustainion ionomer. Sustainion circular micropatterns with diameters of ~300 μm were deposited on GCE (denoted as GCE@Sustainion) through a shadow mask-assisted method (see the Methods section for details, Supplementary Fig. 1). By continuously bubbling CO₂ gas onto the electrode surface, we found that no CO₂ bubbles could be observed on the bare GCE surface (Fig. 1b and Supplementary Fig. 2), but a large number of CO₂ bubbles were captured by the Sustainion micropatterns on GCE@Sustainion (Fig. 1c, d). This indicates the strong affinity of Sustainion ionomer to CO₂, which was further confirmed via the bubble adhesive force measurements. The GCE@Sustainion electrode has a CO₂ adhesive force of 18.2 μN, ~ 3 times higher than that for bare GCE (Fig. 1e). We also performed the real-time bubble adhesion experiments to monitor the dynamic CO₂ transport on the electrode surface using a high-speed camera. The CO₂ bubbles exhibited a steadily pinning state on the GCE surface for up to 100 ms (Fig. 1f), whereas the bubbles quickly spread on the Sustainion ionomer surface in solution within 30 ms (Fig. 1g), indicative of the rapid CO₂ transport on the surface of Sustainion ionomer with a superaerophilic feature. In addition, water contact angle tests showed that the contact angle value increases sharply from 54° for GCE to 122° for GCE@Sustainion (Supplementary Fig. 3), revealing the hydrophobicity of the Sustainion surface, which would help to block the proton transport from the bulk solution to the catalyst surface in acidic electrolyte. The above observations demonstrate that the Sustainion coating layer can effectively facilitate CO₂ capture and transport, and simultaneously restrict the proton transport to catalyst surface, thereby promoting the formation of stable and rich gas–liquid–solid three-phase interfaces for electrocatalytic CO₂R (Fig. 1a).

Synthesis and characterization of catalysts

We further coated the Sustainion ionomer on the surface of Cu catalyst for the study of electrocatalytic CO₂R. We first synthesized a Cu-based metal-organic framework (Cu-BTC) as the Cu precursor by mixing copper nitrate and benzene-1,3,5-tricarboxylate (BTC) in methanol solution at room temperature (Supplementary Fig. 4). Then CuO nanoparticles were obtained through annealing the Cu-BTC in air at 350 °C (Supplementary Fig. 5). Subsequently, CuO nanoparticles and Sustainion ionomer were mixed under ultrasonication and sprayed onto a carbon paper-based gas-diffusion electrode (GDE), which were further subjected to an electrochemical reduction process to form the Cu@Sustainion electrode with the loading amounts of 1.4 mg cm^–2 for Cu and 0.1 mg cm^–2 for Sustainion ionomer (see the Methods section for details).

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images (Supplementary Fig. 6) show that bare Cu has a smooth nanoparticle morphology with diameters of 150–200 nm. After coating of Sustainion ionomer, Cu@Sustainion catalyst exhibits a rough surface with an ionomer layer thickness of 10–30 nm (Fig. 2a, b and Supplementary Fig. 7). X-ray diffraction (XRD) patterns (Supplementary Fig. 8) show that Cu and Cu@Sustainion catalysts have the same diffraction peaks from the cubic Cu phase (JCPDS no. 04-0836). Energy-dispersive X-ray (EDX) elemental mappings (Fig. 2c) revealed that Cl and N elements were uniformly distributed on the Cu surface, further confirming the coating of the Sustainion ionomer layer on the surface of Cu nanoparticles.

Fig. 2: Synthesis and characterization of Cu@Sustainion catalyst.

a–c SEM image (a), TEM image (b), and EDX elemental mappings (c) of Cu@Sustainion catalyst. d FTIR spectra of Cu and Cu@Sustainion catalysts.

The Fourier transform infrared (FTIR) spectrum of Cu@Sustainion (Fig. 2d) shows the absorption bands characteristic of Sustainion ionomer, consistent with the previous report³⁸. From the X-ray photoelectron spectra (XPS) and Auger spectra of Cu element (Supplementary Fig. 9), we can see that bare Cu and Cu@Sustainion catalysts have the same Cu valence states with primary Cu⁰ and minor Cu⁺ species, indicative of the unaffected Cu valence states after the coating of Sustainion ionomer. In addition, the prominent Cl 2p and N 1 s XPS signals from Sustainion ionomer were observed on the Cu@Sustainion catalyst (Supplementary Fig. 9). The CO₂ transport and water contact angle measurements (Supplementary Figs. 10 and 11) revealed that Cu@Sustainion exhibits similar aerophilic and hydrophobic behavior to the above GCE@Sustainion (Fig. 1f, g and Supplementary Fig. 3), indicating the unaltered surface feature of Sustainion ionomer.

Electrocatalytic CO₂R performance in pure CO₂

To verify the predicted enhancement role of the Sustainion coating layer, we first evaluated the electrocatalytic CO₂R performance for Cu@Sustainion and Cu catalysts in a flow cell with pure CO₂ gas fed at the GDE backside (Fig. 3a). A strongly acidic solution containing 3.0 M KCl with 0.05 M H₂SO₄ (pH 0.8) was used as the catholyte^21,22,33. The solution-phase and gas-phase products were quantified via nuclear magnetic resonance (NMR) spectroscopy and online gas chromatography, respectively. The CO₂R product distribution for Cu@Sustainion catalyst (Fig. 3b) shows the increased C₂₊ FE within the current density range from 100 to 800 mA cm^–2. A maximum C₂₊ FE of 78.2 ± 2% with a C₂₊ partial current density of 625.6 ± 16 mA cm^–2 was achieved at 800 mA cm^–2, and the FE for HER was substantially suppressed to below 10%. In contrast, the Cu catalyst displayed a gradual decrease of C₂₊ FE and a gradual increase of HER at current densities of > 300 mA cm^–2 (Fig. 3c). The C₂₊ partial current density for Cu catalyst was merely 305.6 ± 24 mA cm^–2 at a total current density of 800 mA cm^–2, with a C₂₊ FE of 38.2 ± 3% and a H₂ FE of 30.9 ± 4%. Therefore, Cu@Sustainion catalyst exhibits a remarkably enhanced C₂₊ partial current density at high current densities (Fig. 3d). It should be noted that the similar CO₂R activity for Cu and Cu@Sustainion at 100-300 mA cm^–2 could be attributed to the slow consumption and enough supply of CO₂ gas at these low current densities.

Fig. 3: Electrocatalytic CO₂R performance in pure CO₂.

a Schematic illustration of the configuration of a flow cell. Electrolyte: 3.0 M KCl with 0.05 M H₂SO₄ solution (pH 0.8). CO₂ flow rate (pure CO₂): 25 sccm. b, c FEs of all products on Cu@Sustainion (b) and Cu (c) at different current densities. d C₂₊ partial current densities on Cu@Sustainion and Cu. e FEs of C₂₊ products on Cu@Sustainion with different Sustainion loading amounts at 800 mA cm^–2. f FEs of all products and SPCEs on Cu@Sustainion with different CO₂ flow rates at 800 mA cm^–2. The error bars represent the standard deviation of three independent measurements.

We then measured the electrochemically active surface area (ECSA) for Cu@Sustainion and Cu catalysts (Supplementary Fig. 12). Cu@Sustainion catalyst has a smaller ECSA value than Cu catalyst, implying that the enhancement in CO₂R activity for Cu@Sustainion originates from the intrinsic activity rather than the ECSA. We also optimized the loading amount of Sustainion ionomer and found that Cu@Sustainion with a Sustainion loading of 0.1 mg cm^–2 exhibited the highest C₂₊ FE (Fig. 3e and Supplementary Fig. 13). The electrocatalytic CO₂R operated in strongly acidic electrolytes can alleviate the (bi)carbonate formation, thus circumventing the CO₂ utilization limitation typically for alkaline and neutral media^22,26,27. When the CO₂ flow rate decreased from 50 to 2.5 standard cubic centimeters per minute (sccm), Cu@Sustainion catalyst still retained a high C₂₊ FE of ~75%. Therefore, a SPCE of 80.1% for all CO₂R products (~ 61% for C₂₊ products) was achieved at 2.5 sccm and 800 mA cm^–2 (Fig. 3f). Such a SPCE is clearly competitive with those reported in acidic, alkaline, and neutral electrolytes (Supplementary Table 1).

We further evaluated the CO₂R stability for Cu@Sustainion catalyst at a constant current density of 500 mA cm^–2 (Supplementary Fig. 14). The Cu@Sustainion catalyst could retain a high C₂₊ FE of 65–70% for 16 h continuous operation at 500 mA cm^–2. Postmortem characterizations revealed the unchanged morphology and structure for Cu@Sustainion after CO₂ electrolysis (Supplementary Figs. 15 and 16). The above prominent CO₂R activity and stability for Cu@Sustainion could be attributed to the unique Sustainion coating layer, which helps to form stable and rich three-phase interfaces to promote the CO₂R conversion and simultaneously suppress the HER in strong acid.

Moreover, we found that our ionomer coating strategy is quite general. When the Cu component in Cu@Sustainion was replaced by other Cu catalysts, such as commercial Cu nanoparticles (Com-Cu) and electrodeposited Cu microcrystals (ED-Cu), the resulting Sustainion-coated catalysts, denoted as Com-Cu@Sustainion and ED-Cu@Sustainion, respectively, exhibited much improved CO₂R-to-C₂₊ activity and significantly inhibited HER with respect to the uncoated Cu counterparts (Supplementary Figs. 17–20). For example, ED-Cu@Sustainion gives a maximum C₂₊ FE of 76.5% at 800 mA cm^–2, much higher than that for ED-Cu (mere 35.7%) at the same current density (Supplementary Fig. 20).

CO₂R enhancement mechanism

We further conducted multiple characterizations to uncover the origin of the enhanced electrocatalytic CO₂R performance after ionomer coating. The CO₂ concentration at the catalyst surface was determined via an electrochemical method using 1,4-benzoquinone as a redox probe, which can reversibly bind with CO₂ molecules^39,40,41,42. As the local CO₂ concentration can shift the redox potential of 1,4-benzoquinone (Fig. 4a, b), the potential difference (ΔE) in CO₂ and Ar atmospheres for different electrode surfaces can reflect the different local CO₂ concentrations. Cu@Sustainion has a ΔE value of 110 mV, larger than that for Cu catalyst (85 mV), suggesting that the local CO₂ concentration at the catalyst surface increased by ~ 7 times after Sustainion coating (Fig. 4c and Supplementary Table 2 for details). CO₂ gas adsorption measurements reveal that Cu@Sustainion with 6.7 wt% Sustainion displays a CO₂ adsorption capacity 1.7 times higher than Cu (Supplementary Fig. 21), indicating the effective enrichment of CO₂ concentration on the Cu@Sustainion surface.

Fig. 4: Mechanistic investigations.

a, b Cyclic voltammetry curves of 1 mM 1,4-benzoquinone for Cu (a) and Cu@Sustainion (b) in Ar-saturated and CO₂-saturated 0.1 M NaHCO₃ solution. c The potential difference (ΔE) in CO₂ and Ar atmospheres and the derived local CO₂ concentration for Cu and Cu@Sustainion. d, e Linear sweep voltammetry curves with iR correction for Cu (d) and Cu@Sustainion (e) at different electrode rotating speeds in Ar-saturated 3 M KCl with 0.05 M H₂SO₄. f HER kinetic current density (j_k) for Cu and Cu@Sustainion. g, h In situ ATR-SEIRAS spectra for Cu (g) and Cu@Sustainion (h) at different applied potentials.

To verify whether the enhanced CO₂R activity merely originates from the CO₂ enrichment on the hydrophobic chain of Sustainion ionomer, we prepared two control samples by coating the hydrophobic Nafion and polytetrafluoroethylene (PTFE) on Cu surfaces (Supplementary Fig. 22). They could moderately improve the CO₂R activity of Cu, but their activities are still much lower than that for Cu@Sustainion (Supplementary Fig. 23). We further conducted cyclic voltammetry tests and in-situ Raman spectra in organic electrolyte to investigate the role of imidazolium cation (Supplementary Figs. 24 and 25), indicating the strong interaction between electrochemically-generated imidazolium radical and CO₂ molecule. There observations suggest that the hydrophobic chain of Sustainion can enrich CO₂ gas, and the imidazolium cation can chemically interact with CO₂ to enhance the binding ability, which together result in the increased CO₂ adsorption and enrichment near the Cu catalyst surface.

To investigate the effect of Sustainion ionomer on proton diffusion, we conducted rotating disk electrode (RDE) experiments for Cu and Cu@Sustainion in Ar-saturated 3 M KCl with 0.05 M H₂SO₄, where HER is the exclusive reaction. The polarization curves for both electrodes exhibited a current density plateau between − 1.1 V and − 1.4 V, as a result of the HER from diffusion-limiting hydronium reduction (Supplementary Fig. 26)^32,43,44. Compared with Cu, Cu@Sustainion showed smaller plateau current densities at all tested electrode rotating speeds (Fig. 4d, e), indicative of the low local proton concentration with restricted mass transport to the electrode surface. For example, at an electrode rotating speed of 2500 rotations per min (rpm), Cu@Sustainion gave a diffusion-limiting HER plateau current density of ~ 11 mA cm^–2, much lower than that for Cu catalyst (~ 55 mA cm^–2). According to the linear fitting of the Koutecky–Levich equation (Supplementary Fig. 27), we could determine the HER kinetic current density (j_k) decreased from 142.2 to 20.1 mA cm^–2 after Sustainion coating (Fig. 4f), indicating that the Sustainion layer could obstruct the proton diffusion to substantially suppress HER. The ECSA-normalized polarization curves and kinetic current densities further confirmed this observation (Supplementary Fig. 28).

We further performed in-situ attenuated total reflection surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS) to monitor the change of reaction intermediates at different applied voltages during acidic CO₂R (Supplementary Fig. 29). As shown in Fig. 4g, h, the distinct bands at ~ 2030 cm^–1 assigned to *CO intermediates^45,46,47, were observed on both Cu and Cu@Sustainion catalysts. The band intensity for Cu@Sustainion is significantly stronger than that for Cu, indicative of the efficient activation and reduction of the CO₂ molecule to *CO intermediate after Sustainion coating. In addition, the bands at around 1585 cm^–1 were ascribed to the C = O stretching band of *OCCOH intermediates formed via C–C coupling^45,48,49. Compared with Cu catalyst with a weak C = O band started at − 1.1 V, Cu@Sustainion favorably formed this band at a more positive potential and exhibited a stronger band intensity, suggesting that the Sustainion-coated catalyst could readily facilitate the CO dimerization process for the promoted CO₂-to-C₂₊ conversion in acidic electrolyte.

Electrocatalytic CO₂R performance in diluted CO₂

To demonstrate the potential of the direct conversion of flue gas into fuels via CO₂ electrolysis (Fig. 5a), we conducted electrocatalytic CO₂R tests using a simulated flue gas with 15 vol % CO₂ (CO₂/N₂ = 15/85, v/v) as the feed stream at pH 0.8. It is interesting to note that the product FEs for Cu@Sustainion catalyst in diluted CO₂ is close to those in pure CO₂ (Fig. 5b, c and Supplementary Fig. 30). Impressively, Cu@Sustainion can still give a very high C₂₊ FE of 70.5 ± 1.6% at 800 mA cm^–2 in diluted CO₂, only slightly lower than that of 78.2 ± 2% in pure CO₂, thereby retaining 90% of C₂₊ partial current density at 800 mA cm^–2 (Fig. 5e). In stark contrast, when the feed gas was switched from pure CO₂ to diluted CO₂, the C₂₊ partial current density at 800 mA cm^–2 for Cu catalyst decreased markedly from 305.6 ± 24 to 104.2 ± 1 mA cm^–2 with only 34% retention (Fig. 5d, e and Supplementary Fig. 31). Moreover, the SPCE tests revealed that Cu@Sustainion can achieve a high SPCE of 73.6% at a total gas flow rate of 10 sccm containing 15% CO₂ (Fig. 5f), which is comparable to reported benchmark results under pure CO₂ conditions (Supplementary Table 1). These results indicate that Cu@Sustainion could substantially retain the CO₂R performance in diluted CO₂, which can be attributed to the effective enrichment of the diluted CO₂ by the Sustainion layer to generate a high local CO₂ concentration at the catalyst surface, thereby leading to the remarkable CO₂R performance in low-concentration CO₂.

Fig. 5: Electrocatalytic CO₂R performance in diluted CO₂.

a Schematic illustration of the conversion of diluted CO₂ into fuels via CO₂ electrolysis. Electrolyte: 3.0 M KCl with 0.05 M H₂SO₄ solution (pH 0.8). Total gas (CO₂/N₂ = 15/85) flow rate: 50 sccm. b FEs of all products on Cu@Sustainion in diluted CO₂. c FEs of C₂₊ products on Cu@Sustainion in diluted and pure CO₂. d FEs of all products on Cu in diluted CO₂. e Partial current densities of C₂₊ on Cu@Sustainion and Cu in diluted and pure CO₂. f FEs and SPCEs on Cu@Sustainion with different flow rates at 800 mA cm^–2 in diluted CO₂. The flow rate is the total gas flow rate containing CO₂ and N₂ with 15% CO₂. The error bars represent the standard deviation of three independent measurements.

Discussion

In summary, we report an imidazolium-based anion-exchange ionomer-coated Cu catalyst, which can achieve a high C₂₊ FE of 70.5% and a high SPCE of 73.6% at a current density of 800 mA cm^–2 when fed with diluted CO₂ (15 vol% CO₂) at pH 0.8. The CO₂R performance using diluted CO₂ as the feedstock is competitive with results in a pure CO₂ atmosphere. Such good performance could be attributed to the ionomer coating layer that can significantly enrich the diluted CO₂ to generate a high local CO₂ concentration at the catalyst surface, and greatly suppress the HER by obstructing the proton diffusion, thereby forming rich CO₂R three-phase interfaces for effective electroreduction of diluted CO₂ into C₂₊ products in strong acid. Our study provides a promising route to directly converting industrial flue gas into valuable products for energy- and carbon-efficient CO₂R technology.

Methods

Chemicals

Commercial copper nanoparticles (~ 100 nm), Cu(NO₃)₂·3H₂O (99%), CuSO₄·5H₂O (99%), KCl (99.5%), K₂SO₄ (99%), 1,3,5-benzenetricarboxylic acid (BTC, 98%), methanol (99.5%) and isopropanol (99.7%) were purchased from Shanghai Macklin Biochemical Technology Co. Ltd. High purity carbon dioxide gas (99.999%) and nitrogen gas (99.999%) were purchased from Praxair. Nafion 117 membrane was purchased from DuPont. All chemical reagents involved were used without further purification.

Preparation of Sustainion ionomer-coated GCE

The polyimide shadow mask containing circular micropatterns with diameters of 300 μm was fabricated through a direct laser engraving (YLP-HC20, Han’s Laser). Then the polyimide mask was attached on the surface of GCE and Sustainion ionomer (5 wt% in ethanol, XA-9, Dioxide Material) was sprayed onto it. After drying at 60 °C, the polyimide mask was peeled off to form the Sustainion ionomer micropatterns on the GCE.

Preparation of Sustainion ionomer-coated Cu

We used Cu-BTC as the Cu precursor, which was synthesized by mixing Cu (NO)₃·3H₂O (9.1 g) and BTC (0.79 g) in 250 mL methanol. The mixed solution was aged for 6 h at room temperature, followed by centrifugation and drying to obtain Cu-BTC. Then Cu-BTC was annealed at 350 °C for 3 h to obtain the CuO. Subsequently, 7 mg of CuO powder and 8 mg of Sustainion ionomer (5 wt% in ethanol) were dispersed via ultrasonication for 30 min to obtain a homogeneous catalyst ink, which was then sprayed onto a carbon paper-based GDE (YLS 30 T, Toray) with an area of 4 cm². Finally, Cu@Sustainion electrode was obtained through an electrochemical reduction process at a current density of – 20 mA cm^–2 for 30 min in 0.5 M K₂SO₄ electrolyte. The loading amounts of Cu and Sustainion ionomer are 1.4 and 0.1 mg cm^–2, respectively, corresponding to a mass fraction of 6.7 wt% for Sustainion ionomer. Cu@Sustainion electrodes with other Sustainion ionomer loadings of 0.05, 0.2, and 0.5 mg cm^–2 were also prepared by adjusting the ionomer amounts.

For the preparation of Com-Cu@Sustainion, the process was similar with that for Cu@Sustainion except for replacing Cu with Com-Cu (Macklin). For the preparation of ED-Cu@Sustainion, the ED-Cu was first electrodeposited on GDE at – 20 mA cm^–2 for 10 min in 0.1 M CuSO₄ solution and then Sustainion ionomer was sprayed onto the GDE surface to obtain ED-Cu@Sustainion.

Characterization

SEM images were obtained from a scanning electron microscope (FEI Verios 460 L). TEM and HRTEM images were carried out on a transmission electron microscope (FEI Talos F200X). XRD patterns were recorded on an X-ray diffractometer (Japan Rigaku SmartLab 9 kW) with a Cu Kα radiation source (λ = 1.5418 Å). XPS spectra were collected on a Thermo Scientific ESCALAB250Xi spectrometer equipped with an Al Kα (hv = 1486.6 eV) excitation source. FTIR spectra were acquired by using a NICOLET FT-IR spectrometer. CO₂ bubble transport and water contact angle experiments were performed on a dynamic contact angle meter (SZ-CAMC33). CO₂ adhesive force measurements were carried out on a DataPhysics device (DCAT25). In situ ATR-SEIRAS spectra were collected by an FT-IR spectrometer (Nicolet iS50, Thermo Scientific) equipped with an MCT-A detector at different applied potentials.

CO₂R performance measurements and product analysis

We evaluated the CO₂R performance of catalysts in a three-electrode system in a flow cell. Catalyst-coated GDE, Ag/AgCl electrode, IrO₂-coated Pt/Ti mesh were used as the working electrode, reference electrode, and counter electrode, respectively. A strongly acidic solution containing 3.0 M KCl with 0.05 M H₂SO₄ (pH 0.8) was used as the electrolyte, and the electrolyte was used immediately after preparation without storage. The catholyte chamber was separated from the anolyte chamber by a proton exchange membrane (Nafion 117, 183 μm). The catholyte and anolyte were circulated by a peristaltic pump with a flow rate of 20 mL min⁻¹. The cathode was fed with pure CO₂ or diluted CO₂ (CO₂/N₂ = 15/85) controlled by a mass flow controller. In pure CO₂R electrolysis experiments, the gas flow rate is 25 sccm, and other flow rates were used for SPCE tests. In diluted CO₂R electrolysis experiments, the total gas flow rate is 50 sccm containing 15% CO₂, and other flow rates were used for SPCE tests.

The electrochemical results were recorded on an electrochemical workstation (AutoLab PGSTAT 302 N). The FEs of products were measured at a fixed current density. The stability was evaluated via a chronopotentiometry method at a current density of 500 mA cm^–2. The measured potentials versus Ag/AgCl electrode were converted to RHE scale with iR correction according to the following equation: E (vs. RHE) = E (vs. Ag/AgCl) + 0.196 V + 0.059 pH – iR, where R was 5 ± 0.3 Ω from the EIS test.

Gas-phase products were analyzed on an on-line gas chromatography system (Nexis GC-2030), and liquid-phase products were quantified on an NMR spectrometer (Bruker AVANCE III 400 MHz).

The FE of the products was calculated using the following method:

$${{{{\rm{FE}}}}}_{{{{\rm{gas}}}}}=\frac{z\times {{{\rm{F}}}}\times \upsilon \times r}{i\times {V}_{{{{\rm{m}}}}}}\times 100\%\,{{{\rm{and}}}}\,{{{{\rm{FE}}}}}_{{{{\rm{liquid}}}}}=\frac{z\times {{{\rm{F}}}}\times n}{{{{\rm{Q}}}}}\times 100\%$$

Where z is the electron transfer number, F is Faraday’s constant (96,485 C mol⁻¹), υ is the gas flow rate measured at the reactor outlet (L min⁻¹), r is the detected gas product concentration in parts per million (ppm), i is the total applied current (A), V_m is the molar gas volume (24.5 L mol⁻¹), n is the moles of liquid products from NMR analysis (mol), and Q is the total charge (C).

The SPCE of CO₂R was calculated according to the following equation:

$${{{\rm{SPCE}}}}=\frac{60{{{\rm{s}}}}\times \sum (i\times {{{{\rm{FE}}}}}_{{{{\rm{i}}}}}\times {x}_{i}\div(z\times {{{\rm{F}}}}))}{y\times {{{\rm{flow\; rate}}}}\,({{{\rm{L}}}}/\min )\times 1{{\mathrm{min}}}/24.5({{{\rm{L}}}}/{{{\rm{mol}}}})}$$

where i is the total applied current (A), FE_i is the Faradaic efficiency for each CO₂R product, x_i is the molar ratio of CO₂ to each CO₂R product (such as x = 1 for C₁ product, x = 2 for C₂ product, and x = 3 for C₃ product), z is the electron transfer number for each CO₂R product, F is Faraday’s constant, and y is volume fraction of CO₂ (such as y = 100% for pure CO₂ and y = 15% for diluted CO₂).