2.1. Electrodeposition of Zn-Based Catalyst

The Zn-based catalyst for the CO₂ electrolysis was fabricated by a facile electrodeposition method. A home-made Teflon cell with 1.44 cm² of exposed area was used as the three-electrode cell system and was connected with a potentiostat (Autolab, PGSTAT302N, Metrohm). The working electrode had a microporous layer on a carbon paper (MPL/CP). A saturated calomel electrode (SCE) and a Pt wire were used as the reference and counter electrodes, respectively.

Prior to the electrodeposition, the MPL/CP surface was electrochemically pretreated at 5 mA/cm² for 100 s in a 0.5 M (NH₄)₂SO₄ solution to improve the surface hydrophilicity. The deposition electrolyte was composed of 0.2 M ZnSO₄·7H₂O, 0.5 M (NH₄)₂SO₄, and 1.0 M H₂SO₄. The electrodeposition for the Zn/MPL/CP fabrication was conducted by applying a constant potential at –2.1 V_SCE until the deposited charge amount was leached at –50 C. The deposition solution was N₂-purged before the electrodeposition step to remove the dissolved O₂ species.

2.2. Ionomer Coating on As-Prepared Zn/MPL/CP

The ionomer was spray coated on the as-prepared Zn/MPL/CP to further improve the CO₂ electrolysis performance. The ionomer solution consisted of 736 μL of deionized water (DI), 64 μL of the ionomer (XA-9, Dioxide Materials), and 3.2 mL of isopropyl alcohol (IPA). The ionomer-coated Zn/MPL/CP cathode was dried at 50°C for 30 min to evaporate the residual IPA.

2.3. Characterization

Field-emission scanning electron microscopy (FESEM, Sigma, Carl Zeiss) and energy-dispersive X-ray spectroscopy (EDS, Thermo, NORAN System 7) analyses were conducted to investigate the surface morphology and bulk atomic composition of the as-prepared Zn/MPL/CP. The bulk crystal structure was also observed through X-ray diffraction (XRD, Bruker, D8 Advance). The electronic structure and surface composition were analyzed by X-ray photoelectron spectroscopy (XPS, K-alpha+, Thermo Fisher Scientific). In addition, the Raman (LabRAM HR Evolution, HORIBA) analysis was conducted to confirm the oxidation state of the Zn/MPL/CP surface.

2.4. Electrochemical Measurements in the Conventional H-Type Electrolysis System

The CO₂ electrolysis performance in an H-type cell system was verified for 20 min at each constant potential. The as-prepared Zn/MPL/CP was used as the working electrode in an H-type electrolyzer. Further, Pt-mesh and SCE were used as the counter electrode and reference electrode, respectively. The working and counter electrodes were separated by a membrane, and Nafion 212 (Dupont Co.) and Sustainion X37-50 (Dioxide Materials) were used for the proton exchange membrane- (PEM-) based H-type electrolyzer and AEM-based H-type electrolyzer, respectively. The active area for electrodeposition was 1.0 cm².

The CO₂-purged and N₂-purged 0.5 M KHCO₃ electrolytes were contained in the cathode and anode, respectively. Then, the CO₂ was continuously injected with a flow rate of 10 mL/min by modulating the mass flow controller (MKS Instruments, Inc.). The measured potential was converted to a reversible hydrogen electrode (RHE) with the following formula: E_RHE = E_SCE + 0.2412 + 0.0592 × pH. The outlet of each electrolyzer was connected to a gas chromatograph (GC, Agilent 7890B) to measure the concentrations of the gas products (H₂ and CO).

The gas products from the electrolyzer were gained for 1 min before the end of each potential measurement, and simultaneously, the flow rate of the outlet was detected by the mass flow meter. The Faradaic efficiency (FE) for each product was calculated by integrating the concentration from GC, flow rate, and the charge amount during the time of product gained. The partial current density (PCD) which represents the current density for a specific product obtained was calculated by multiplying the total current density with FE.

2.5. Electrochemical Measurements in the MEA-Type Electrolysis System

Unlike the conventional H-type cell system, the MEA-type electrolyzer was connected with a two-electrode cell system. The Zn/MPL/CP was used as the cathode, and the commercial IrO₂/CP (C0206 Dioxide Materials) was used as the anode. In the case of the MEA-type electrolyzer, the membrane (PEM: Nafion 212, Dufont Co., AEM: sustainion x37-50, Dioxide Materials) was located between the cathode and anode with a zero-gap configuration, where the active area was 1.0 cm².

Humidified CO₂ gas was supplied into the backside of the Zn/MPL/CP cathode with a flow rate of 200 mL/min. On the anode side, DI and 0.1 M KOH were injected with a flow rate of 20 mL/min for the PEMEA and AEMEA-type electrolyzers, respectively. The FE and PCD for CO and H₂ from the MEA-based electrolyzer were calculated in the same manner as the H-type electrolyzer.

3.1. Characterization of the Instability of the Zn-Based Catalyst

Prior to testing the CO₂ electrolysis performance, we investigated the unstable properties of the Zn-based catalyst under an aqueous electrolyte. The Zn/MPL/CP was fabricated by electrodeposition on the MPL/CP substrate (Figure S1), and the FESEM image shows that the Zn deposits with the hexagonal structure were well distributed (Figure 1(b)). However, when the as-prepared Zn/MPL/CP was immersed into a 0.5 M KHCO₃ electrolyte, the hexagonal structure gradually changed to a dendrite (Figure 1). This morphological change in the Zn-based catalyst was due to the difference in the dissolution rates based on the crystal structures, which was confirmed in our previous study [37]. The conversion to the dendrite structure was saturated at an immersion time of 180 min, which indicates the unstable properties of the Zn materials in the electrolytes.

Figure 2 shows the comparison of the properties of the Zn/MPL/CP before and after immersion into the KHCO₃ electrolyte. The structural change in the Zn/MPL/CP was investigated by XRD analysis (Figure 2(a)). For the as-prepared Zn/MPL/CP, three peaks were detected in the XRD pattern (JCPDS #87-0713): Zn(002) at 36.3°, Zn(100) at 39.0°, and Zn(101) at 43.2°, indicating that the as-prepared Zn catalyst was close to the metallic state. In addition, the Zn(101) facet was dominant in the as-prepared Zn/MPL/CP, indicating that it was suitable for CO production via CO₂ electrolysis [13, 38]. After immersion into the 0.5 M KHCO₃ solution for 180 min, the new peak appeared at 31.8°, corresponding to the ZnO(100) facet.

As shown in Figure 2(b), compared to the Zn/MPL/CP, the ZnO(100)/Zn(100) intensity ratio increased in the I180-Zn/MPL/CP where the immersed time was 180 min. The additional results of the XRD patterns of the Zn/MPL/CP with varying immersion times are presented in Figure S2a and b. The XRD analysis revealed that a direct contact between Zn and the electrolyte caused not only a morphological change but also a chemical change from a metallic to an oxidized state.

Ex situ Raman spectroscopy was conducted to investigate the change in the surface structure of the Zn/MPL/CP after immersion for 180 min. As shown in Figure 2(c), two peaks at 387 cm^–1 and 560 cm^–1 were observed, which correspond to the characteristic vibrational mode of ZnO [12, 39]. In the case of the I180-Zn/MPL/CP, an increase in the peak intensity was observed at 387 cm^–1, 560 cm^–1, and ~240 cm^–1, which revealed that these are ascribed to the characteristic vibrational mode of Zn(OH)₂ [40].

Further, we demonstrated the electronic structures of the surfaces of the Zn/MPL/CP and I180-Zn/MPL/CP using the XPS analysis. As shown in Figure 2(d), the XPS spectrum of the Zn 2p_3/2 region was deconvoluted into three peaks: 1021.3 eV, 1021.8 eV, and 1023.0 eV, which were assigned to metallic Zn, Zn-O, and Zn-OH, respectively [41]. After the immersion into the 0.5 M KHCO₃, the intensity of the metallic Zn was reduced, whereas those of the Zn-O and Zn-OH states increased. Meanwhile, three separated peaks were observed at O 1 s: 530.1 eV, 531.4 eV, and 532.2 eV, which were assigned to Zn-O, Zn-OH, and absorbed water molecules, respectively (Figure 2(e)) [41]. It was revealed that the ratio of Zn-OH and water molecules in the O 1 s peak increased after the immersion.

In addition, a significantly higher intensity at 376.6 eV representing the K 2 s state was observed in the I180-Zn/MPL/CP, which could be due to potassium incorporation into the surface (Figure 2(f)). The increase in the atomic compositions for O and K with the increasing immersion time was also confirmed by the EDS analysis (Figure S2c). The above results demonstrate that a direct contact between unstable Zn material and electrolyte causes surface oxidation and morphological changes.

3.2. Electrochemical Properties of the Zn/MPL/CP according to the Electrolyzer Type

Although various studies have reported on the high performance of the Zn-based catalysts in CO₂ electrolysis, the stability issues of the electrocatalysts still remain unaddressed [11, 42]. In the conventional H-type cell for the CO₂ electrolysis, a direct contact between the electrode and electrolyte is inevitable, leading to performance degradation. To investigate the corrosion resistance of the Zn/MPL/CP as a cathode according to the electrolyzer configuration, the cell voltage variation in the two-electrode cell system was measured during the open-circuit period (Figure 3 and Figure S3).

In the conventional PEM-based H-type electrolyzer system, the measured cell voltage was started at –1.47 V_cell; however, the value decreased dramatically to –1.6 V_cell over 10 min and was then saturated (Figure 3(a)). Furthermore, the hexagonal structure of the Zn/MPL/CP was agglomerated and transformed into a dendrite after the measurement (Figure 3(d)), which corresponds to the FESEM images in Figure 1.

On the other hand, in the case of the PEMEA-type electrolyzer, a smooth decrease in the cell voltage from –1.25 V_cell to –1.36 V_cell within 60 min was observed. The hexagonal Zn structure was well preserved after the test, although there were weak surface damages (Figure 3(c)). Similar trends were also observed in the AEMEA-type electrolyzer, as shown in Figure 3(b). The EDS analysis after the measurement revealed that the metallic Zn state of the Zn/MPL/CP was well maintained when using the MEA-type electrolyzer (Figure S4). It means that the presence of a polymer membrane can mitigate the chemical oxidation of the Zn-based catalyst surface [43]. However, it should be noted that the precise comparison of the cell voltage variation was difficult because the ion migration differed according to the membrane and electrolyte used. Nevertheless, it implies that using MEA-type electrolyzer is essential not only for enhancing the CO₂ electrolysis performance but also for achieving durability.

3.3. Measurement of CO₂ Electrolysis Performance according to the Electrolyzer Type

We investigated the CO₂ electrolysis performance for 180 min at a constant potential according to the electrolyzer used (Figure 4). In the case of the PEM-based H-type electrolyzer, the Zn/MPL/CP cathode exhibited relatively low activity for the CO production at –1.0 V_RHE due to the poor solubility of CO₂ into the aqueous electrolyte (Figure 4(a)). Furthermore, the highest FE for the CO production was measured as 63% at 60 min, which steadily deteriorated with the increasing operation time. In addition to low CO₂ solubility, the ohmic loss owing to the presence of an electrolyte layer contributed to low CO₂ electrolysis performance.

In the case of PEMEA-type electrolyzer, hundreds of total current densities at –2.8 V_cell were obtained and gradually increased during the reaction (Figure 4(b)). It was because of the MEA configuration leading to improve energy efficiency and facilitate the mass transport. However, the calculated FE confirmed the most of the current density contributed to the hydrogen evolution reaction (HER) which is competing reaction against electrochemical CO₂ reduction. It was that the H⁺ generated via the oxygen evolution reaction at the anode was transferred to the cathode through the PEM. As a result, the H⁺ local concentration near the cathode surface was relatively higher than that of the CO₂ reactant.

On the other hand, the current density for CO production via CO₂ electrolysis was further enhanced in the AEMEA-type electrolyzer (Figure 4(c)). This indicated that the AEM could suppress the accumulation of H⁺ near the cathode mitigating the HER [44, 45]. Accordingly, the CO₂ electrolysis was dominant at the cathode and the AEMEA-type electrolyzer exhibited ~76% of stable CO FE for 180 min. Although there was slight performance degradation during the 180-min operation, the CO FE was well maintained, implying that the catalytic properties were unchanged.

The surface morphology of the Zn/MPL/CP cathode after the CO₂ electrolysis according to electrolyzer configuration was investigated by FESEM, which indicated that the Zn electrocatalysts were agglomerated with deformation in an PEM-based H-type system (Figure S5a). On the other hand, a damaged hexagonal surface was observed in the PEMEA-type electrolyzer due to the high H⁺ concentration near the surface (Figure S5b).

In the case of the AEMEA-type electrolyzer, the hexagonal fixture of the Zn/MPL/CP was well maintained after 180 min of CO₂ electrolysis (Figure S5c). Furthermore, the AEMEA-type system exhibits the higher CO selectivity with high PCD at various applied potentials than other systems (Figure S6). This indicates that not only the MEA configuration, particularly when using AEM, but also the direct supply of gaseous CO₂ could provide reasonable CO₂ electrolysis performance while preserving the catalyst properties. The low K atomic ratio observed by the EDS analysis after the CO₂ electrolysis also indicated that the contribution of the catholyte in the AEM-based electrolyzer was reduced (Figure S5d).

3.4. Ionomer Effects on the AEMEA-Type CO₂ Electrolysis

To further enhance the CO₂ electrolysis performance, we introduced AEI layer on the top surface of the Zn/MPL/CP by spray-coating. The AEI layer provides an ion pathway and improves the interface properties between the electrode and membrane that can maximize the reaction active sites. The amount of AEI content was controlled by repeating the spray-coating process.

Figures 5(a), 5(b), and 5(c) show the FESEM images of the ionomer-coated Zn/MPL/CP cathode according to deposited AEI amounts. Compared to the as-prepared Zn/MPL/CP cathode (Figure 1(b)), the thin film was formed on the surface of the Zn/MPL/CP cathode containing 90.2 μg/cm² of the AEI, and the hexagonal structure was well preserved (Figure 5(a)). As the AEI content increased to 111.1 μg/cm², the hexagonal Zn interconnected with each other (Figure 5(b)), and an excessive amount of AEI (234.1 μg/cm²) blocked the CO₂ gas passage (Figure 5(c)). Meanwhile, the EDS analysis indicated that with the increasing amount of AEI content, which consisted of chloride form, the Cl ratio increased (Figure 5(d)) [45].

The AEMEA-type CO₂ electrolysis performance for the AEI-coated Zn/MPL/CP cathodes was measured at –2.8 V_cell for 20 min (Figure 6). The CO FE exhibited an increasing tendency with the increasing amount of AEI, as shown in Figure 6(a). Correspondingly, the CO PCD improved from –68.1 mA/cm² to –107.8 mA/cm² (Figure 6(a)). However, the excessive amount of AEI coating (234.1 μg/cm²) led to degradation of the CO PCD to –92.8 mA/cm² because of blockage in the pore structure and reduced active sites, as shown in Figure 5(c).

Meanwhile, the H₂ FE and H₂ PCD decreased after the AEI coating on the Zn/MPL/CP surface. It could be that the ionomer thin layer mitigates the accumulation of H⁺ and allows the enhancement of CO selectivity. In addition, during the AEMEA-type CO₂ electrolysis at –2.8 V_cell for 180 min, the AEI-coated Zn/MPL/CP cathode exhibited higher CO FE and CO PCD than those of the bare Zn/MPL/CP cathode. The additional electrochemical performance of AEI-coated Zn/MPL/CP according to the operating time is shown in Figure S7.

Moreover, the FESEM images in Figure S8 confirmed the hexagonal structure of the Zn/MPL/CP with the AEI content was well preserved after the CO₂ electrolysis for 180 min. This implied that the AEI thin layer could mitigate the access of the electrolyte or cations from the anode to the cathode, thus contributing to HER suppression. Additionally, the stability test for 12 h was conducted, and although there was a slight degradation in current density, the CO FE was well maintained at ~75% (Figure S9). It should be noted that we focused on systemic approaches for enhancing the performance and stability of CO₂ electrolysis with unstable Zn-based materials in this study; therefore, further development of a state-or-the-art electrocatalyst or an electrode with high performance is necessary to improve the overall system efficiency.