Decoupling‐Facilitated Mass‐Charge Transfer via Dual‐Interface Engineering for Efficient CO₂ Electrolysis

ABSTRACT

The gas‐electrolyte‐electrode triple‐phase interfaces (TPIs) critically govern the kinetics of the electrochemical CO₂ reduction reaction (CO₂RR) by regulating concerted proton‐electron transfer processes. However, sluggish mass transfer and the unbalanced adsorption of key intermediates within the local microenvironment of TPI remain major obstacles to efficient multicarbon product formation. Here, we report a dual‐interface strategy featuring amphiphilic and biphasic architectures to decouple mass‐charge transfer, achieved through in‐situ electrochemical activation of a polydimethylsiloxane (PDMS)‐modified Cu‐BTC electrode. The hydrophilic/hydrophobic interface promotes the synergistic mass transfer of CO₂ and protons within the TPI microenvironment, whereas the amorphous/crystalline interface modulates the electronic structure of catalytic active sites to optimize the adsorption kinetics of key intermediates. Such decoupling‐mediation accelerated C₂H₄ Faradaic efficiency (FE) exceeding 86% at ‐0.9 V (vs. reversible hydrogen electrode, RHE), over 2.5 times higher than that of the control groups. This work highlights the potential of dual‐interface decoupling engineering to simultaneously optimize CO₂ mass transport pathways and intermediate adsorption kinetics, thereby enabling highly efficient electrosynthesis of C₂H₄.

A dual‐interface strategy is constructed by in‐situ activations during electrochemical CO₂ reduction to boost localized mass transfer and charge transfer. Hence, the interfaces between the amorphous phase and the hydrophilic Cu⁰/Cu⁺ crystalline phase in the hydrophobic microenvironment realize decoupled optimization of key intermediate adsorption and promote C‐C coupling, enabling highly selective and stable ethylene production.

1. Introduction

Utilizing renewable electricity to drive the electrochemical CO₂ reduction reaction (CO₂RR) is essential for achieving carbon neutrality and generating high‐value feedstocks [1, 2, 3, 4]. Ethylene (C₂H₄) is particularly attractive due to its high energy density and economic value [5, 6, 7, 8]. However, the current CO₂RR is still constrained by three major bottlenecks: insufficient CO₂ mass transfer efficiency, imbalanced adsorption energy barriers of key intermediates, and sluggish charge transfer kinetics, which particularly affect the directed synthesis of multi‐carbon products such as C₂H₄ [9, 10, 11, 12].

Design of triple‐phase interfaces (TPIs) and local microenvironments is promising to enhance CO₂ mass transport from the electrolyte to the catalyst surface [13, 14, 15, 16]. The gas‐electrolyte‐electrode TPI is the core reaction region in electrochemical CO₂RR, as it serves as the primary reaction zone where gaseous CO₂, solvated ions, and electrons simultaneously converge [17, 18, 19]. However, the intrinsic coupling between CO₂ mass transport, proton availability, and electron transfer at this interface often leads to severe kinetic limitations and competing hydrogen evolution reaction (HER) [20, 21, 22]. Such coupling becomes particularly detrimental for C₂H₄ formation, which requires both high local ^*CO coverage and a well‐regulated reaction microenvironment for efficient C‐C coupling. Therefore, decoupling these interfacial processes and constructing well‐defined dual interfaces represents a promising strategy to optimize reaction pathways and enhance CO₂RR selectivity toward C₂H₄ production [23, 24, 25].

Here, we propose a dual‐interface strategy decoupling mass‐charge transfer by constructing local dual interfaces through electrochemical in‐situ activation, including hydrophilic‐hydrophobic interfaces and amorphous/crystalline interfaces. Specifically, the polydimethylsiloxane (PDMS)‐modified Cu‐BTC electrode was activated in situ by applying a potential to obtain PDMS@Cu‐4 with local dual interfaces. PDMS can not only regulate the local hydrophilic‐hydrophobic environment on the electrode surface and optimize the mass transfer process of CO₂ and protons, but also act as a protective layer to stabilize the catalytic active sites and inhibit the corrosion and deactivation of the catalyst. Strikingly, the PDMS@Cu‐4 catalyst with decoupling‐mediation dual interfaces achieved an FE_C2H4 of 86% at a low overpotential and exhibited 24 h operational stability. Meanwhile, the FE_C2H4/FE_H2 ratio reached 25, which is about 2.5 times that of other similar catalysts. Consequently, this study provides a new interface regulation idea and experimental basis for the efficient production of C₂H₄ products by electrocatalytic CO₂RR.

2. Results and Discussion

2.1. Characterization of Dual Interface

The in situ activated PDMS@Cu‐4 was obtained by electroactivation at −1.3 V for 30 min along with CO₂RR. During such an in situ self‐adaptation process, it adaptively constructs hydrophilic/hydrophobic local interfaces and amorphous/crystalline composite interfaces (Figure 1a). The crystal structures of PDMS@Cu‐0 and PDMS@Cu‐4 were accurately determined using high‐resolution transmission electron microscopy (HRTEM) imaging technology. PDMS@Cu‐0 shows interplanar spacings of 0.178 nm and 0.208 nm, corresponding to the CuO(112) and Cu₂O(200) planes, respectively. Cross‐lattices in the circled magnified image (red dashed lines) show interplanar spacings of 0.18 nm and 0.174 nm, corresponding to the Cu(200) and CuO(112) planes, respectively. It is proved that there are Cu(0), Cu(I), and Cu(II) sites in PDMS@Cu‐0 (Figure 1b). PDMS@Cu‐4 shows interplanar spacings of 0.203 nm and 0.243 nm, corresponding to the Cu(111) and Cu₂O(111) planes, respectively (Figure 1e). These interplanar spacings indicate the presence of Cu(0) and Cu(I) sites in PDMS@Cu‐4.

FIGURE 1.

Schematic of design concept and structural characterizations. (a) Schematic diagram of dual interface design concept. (b, c) HRTEM images of PDMS@Cu‐0. (d) The upper shows the CA of PDMS@Cu‐0, and the lower shows the CA of PDMS@Cu‐4. (e) HRTEM image of PDMS@Cu‐4. (f) HAADF‐STEM image of PDMS@Cu‐4. (g) The elemental mapping images of PDMS@Cu‐4. The scale is 50 nm. (h,i) Operando XANES (h) and FT‐EXAFS (i) spectra of the Cu K‐edge recorded on PDMS@Cu‐4 at different applied potentials, from open‐circuit potential (OCP) to −1.5 V of electrocatalytic CO₂RR. Unless otherwise stated, all potentials are referred to the reversible hydrogen electrode (RHE).

Meanwhile, HRTEM image analysis of PDMS@Cu‐0 and PDMS@Cu‐4 reveals distinct short‐range order coexisting with amorphous domains (Figure 1b,c,e). The formation of these amorphous regions is attributed to PDMS encapsulation, a hypothesis strongly supported by energy‐dispersive X‐ray spectroscopy (EDS) mapping. This structural feature is central to our strategy for dual interface regulation. In the precursor PDMS@Cu‐0, elemental mapping reveals homogeneous silicon (Si) dispersion across the Cu‐BTC surfaces (Figure S1), confirming successful PDMS infiltration into the Cu‐BTC architecture [20]. This structure is a prerequisite for establishing the desired local hydrophilic/hydrophobic environment to optimize CO₂ mass transport. Crucially, in the activated catalyst PDMS@Cu‐4, EDS mapping demonstrates Si elements spatially correlated with copper centers (Figure 1f,g), indicating intimate contact and the formation of the amorphous/crystalline phase interface between PDMS and the reconstituted Cu species. This multiscale structure unambiguously confirms the amorphous phase as PDMS‐derived, directly demonstrating the successful realization of amorphous/crystalline phase interfaces through our electrochemical activation strategy.

Static contact angle (CA) measurements reveal PDMS‐driven interfacial engineering (Figure 1d). PDMS@Cu‐0 exhibits pronounced hydrophobicity (143°), significantly exceeding pristine Cu‐BTC (129°, Figure S2), confirming enhanced surface encapsulation. Notably, PDMS@Cu‐4 transforms to a hydrophilic state (83°), demonstrating voltage‐induced amphiphilic restructuring. This dynamic wettability transition enables simultaneous CO₂ enrichment (hydrophobic domains) and ion hydration (hydrophilic channels), optimizing triple‐phase reactivity.

Operando X‐ray absorption near‐edge structure (XANES) and extended X‐ray absorption fine structure (EXAFS) were used to further investigate the local coordination environment and electronic structure of the PDMS@Cu‐4 catalyst. The Cu foil, Cu₂O Ref., and CuO Ref. were also tested as a comparison to PDMS@Cu‐4. During electrochemical reduction, the valence state of PDMS@Cu‐4 progressively reduced to Cu^δ+ (0< δ < 1) with increasing applied voltage (Figure 1h; Figure S3) [25, 26]. Concurrently, PDMS@Cu‐4 exhibited prominent unsaturated Cu‐O coordination. Post‐electroreduction characterization revealed a controlled reduction of Cu─O─Cu bonds to metallic Cu species, accompanied by a significant increase in Cu─Cu bond content.

It is worth noting that the strength of Cu─Cu bonds becomes weaker from −0.7 V to −1.5 V, but is much higher than OCP (Figure 1i,j; Figure S4). This unique structural evolution is a direct consequence of the amorphous PDMS encapsulation, which kinetically stabilizes specific copper oxide domains against complete reduction [27]. The stabilization of this amorphous PDMS/crystalline Cu^δ+ composite structure is paramount. These phenomena demonstrate that the amorphous PDMS intimately integrated with the crystalline Cu^δ+ modulates the electronic structure of the catalytic copper sites. This electronic modulation engineered by the amorphous/crystalline phase interface is essential for optimizing the adsorption energy barrier of critical intermediates like ^*CO, thereby promoting the kinetically challenging C–C coupling step crucial for C₂H₄ formation [28]. This ensemble of results unequivocally demonstrates that the PDMS encapsulation orchestrates a dynamic in‐situ activation process, precisely constructing and stabilizing the crucial amorphous/crystalline phase interface. As mentioned above, operando XAS studies combined with TEM and CA analysis demonstrate that PDMS@Cu‐4 simultaneously forms local hydrophilic/hydrophobic interfaces and amorphous/crystalline composite interfaces following PDMS modification and electrochemical in situ activation.

2.2. Mechanism Investigations and DFT Analysis

To understand the mechanism of PDMS@Cu‐4 in CO₂RR at the molecular level and to elucidate the origin of its excellent product selectivity, we performed in situ attenuated total reflection Fourier transform infrared spectroscopy (ATR‐FTIR) studies [29, 30]. The test was carried out in a self‐designed electrolytic cell (Figure S5). During the test, a potential of −0.9 V was applied to the PDMS@Cu‐4 catalyst for 60 min to track species evolution characteristics during CO₂RR. As time increases, in situ ATR‐FTIR spectra and corresponding contour plots (Figure 2a,b) show that progressively enhanced peaks are detected at spectral bands 1868, 1651, and 1423 cm⁻¹, respectively, indicating the coverage of the generated ^*CO, adsorbed H₂O, and ^*COOH intermediates [31, 32]. Simultaneously, peak signals from ^*COCHO (1260 cm⁻¹), ^*COCO (1539 cm⁻¹) and ^*CHO (1740 cm⁻¹) increased steadily [28, 33]. These results suggest that increased ^*CO coverage on the PDMS@Cu‐4 surface controls the formation pathway of the C₂H₄ products, avoiding continuous protonation of ^*CO and thus promoting C─C coupling to form ^*COCHO after a single protonation step.

FIGURE 2.

Mechanism investigations. (a, b) In situ ATR‐FTIR spectra (a) and corresponding contour plot of PDMS@Cu‐4 recorded during CO₂RR at −0.9 V, capturing surface intermediate changes from 0 to 60 min. (c,d) In situ Raman spectra (left) and corresponding contour plot (right) of PDMS@Cu‐0 (c) and PDMS@Cu‐4 (d) during CO₂RR at different potentials from OCP to −1.0 V in 0.5 m KHCO₃. (e–h) Differential charge analysis of ^*CO absorbed by Cu(I)/Cu(0)‐PDMS surface (front view: e, top view: f) and Cu(0)‐PDMS surface (front view: g, top view: h). The blue, red, brown, gray, and pink balls represent copper, oxygen, carbon, silicon, and hydrogen atoms, respectively. The yellow and blue iso surfaces correspond to the electron accumulation region and depletion region, respectively. (i,j) The PDOSs of Cu(0)/Cu(I)‐PDMS (i) and Cu(0)‐PDMS (j). E_F is the Fermi level. (k) The energy diagram shows the key intermediates of Cu(0)/Cu(I)‐PDMS and Cu(0)‐PDMS leading to C₂H₄ production.

To further investigate the chemical properties of PDMS@Cu‐0 and PDMS@Cu‐4 and the adsorption of intermediates, in situ Raman spectroscopy measurements were performed during CO₂RR using a self‐designed electrolytic cell (Figure S6). As shown in Figure 2c,d, the vibration of the C─H bonds was observed in the range of 2700 to 3000 cm⁻¹ [34]. The peak at 2908 cm⁻¹ was detected on both PDMS@Cu‐0 and PDMS@Cu‐4, and the peak on PDMS@Cu‐4 was more obvious than that on PDMS@Cu‐0. It is worth noting that PDMS@Cu‐4 also has a peak at 2967 cm⁻¹. This indicates that the surface number of reaction intermediates containing C─H bonds (such as ^*CHO and ^*C₂H₂O) on PDMS@Cu‐4 is higher [35]. These C─H‐containing intermediates are essential for the formation of C₂H₄ products [34, 35, 36]. It can enhance the coverage of surface ^*CO and promote the pathway of ^*CO dimerization to form ^*COCHO, thereby improving the selectivity of C₂H₄ during CO₂RR.

In order to investigate the adsorption of CO₂, ^*CO, and ^*H on PDMS@Cu‐4, we carried out density functional theory (DFT) analysis. Cu(111) surface, Cu₂O(111) surface and Cu₂O(111)/Cu(111) interface were selected as catalytic active site models. First, the adsorption energies of CO₂, ^*CO, and ^*H at the interfacial region of the catalyst were compared (Figure S7) [24]. The Cu(0)/Cu(I) site (−1.56 eV) exhibited a more favorable CO₂ adsorption than the Cu(I) site (−0.34 eV) and the Cu(0) site (−0.20 eV), thus providing an activation pathway for CO₂RR. On the contrary, the order of ^*CO adsorption from strongest to weakest was Cu(0) > Cu(0)/Cu(I) > Cu(I), suggesting that the Cu(0) site helps to prevent the desorption of ^*CO from the surface of the catalyst, thus increasing the surface coverage of ^*CO for the subsequent C‐C coupling step [24, 28]. In addition, the free energy of ^*H at the Cu(0)/Cu(I) sites is higher than that of both Cu and Cu₂O, supporting the effective inhibition of HER at the interface.

Differential charge analysis (DCA) was conducted to explore the electronic coupling between the ^*CO intermediate and the biphasic Cu(0)/Cu(I)‐PDMS interface (represented by Cu(111) and Cu₂O(111)). This was compared with a reference system of pristine Cu(0)‐PDMS (represented by Cu(111)) to clarify the influence of the biphasic structure on charge redistribution [6]. Figure 2e–h illustrates the electron transfer from the two Cu surfaces to the ^*CO intermediates. Bader charge analysis reveals that ^*CO adsorbed on the Cu(0)/Cu(I)‐PDMS surface gains only 0.14 electrons, indicating weaker electronic interaction compared to that on the Cu(0)‐PDMS surface, which gains 0.35 electrons. This suggests that the Cu(0) site helps prevent ^*CO from desorbing from the catalyst surface, thereby increasing the surface coverage of ^*CO to facilitate the subsequent C‐C coupling step. This is consistent with in situ ATR‐FTIR experimental results. Based on the calculations of projected density of states (PDOS) and the analysis of the energy band center of Cu atom 3d orbitals, it is found that the d‐band center of Cu atoms in Cu(0)/Cu(I)‐PDMS (−1.98 eV) is more positive (closer to the Fermi level) than that in Cu(0)‐PDMS (−2.21 eV). This electronic structure feature results in a stronger binding interaction between Cu atoms and the ^*CO intermediate, and such an enhanced binding is expected to improve the CO₂RR catalytic activity of Cu(0)/Cu(I)‐PDMS. It is noteworthy that the O 2p orbital maintains a good overlap with the Cu 3d orbital, which contributes to the good stability of Cu(0)/Cu(I)‐PDMS (Figure 2i,j).

To reveal the effect of the dual interface in CO₂RR, we calculated the Gibbs free energies of key intermediates involved in C₂H₄ generation during CO₂RR over Cu(0)‐PDMS and Cu(0)/Cu(I)‐PDMS catalysts [28]. Optimized configurations of ^*CO₂, ^*COOH, ^*CO, ^*CO+^*CO, and ^*COCO intermediates adsorbed on the surfaces of these two catalysts are shown in Figures S8 and S9. The adsorption and activation of CO₂ molecules on the catalyst surface represent the initial step of CO₂ conversion. Here, ^*CO₂ was used as the starting point for calculations, with the adsorption capacity of the catalyst for CO₂ not considered in the model. As confirmed in Figure 2k, the formation of ^*COOH on the Cu(0)/Cu(I)‐PDMS surface is significantly easier than on the Cu(0)‐PDMS surface [6, 37]. This facilitates more efficient generation of ^*CO, indicating that dynamic Cu^δ+ species formed during the reaction induce more effective stabilization of ^*CO intermediates, which is directly related to the decoupling‐mediation of the dual interface.

2.3. Morphological and Structural Characterizations of Catalysts

As shown in Figure 3a, the scanning electron microscopy (SEM) images clearly show the significant evolution of the morphology of PDMS@Cu‐0 as the applied voltage increases. When no voltage is applied for activation, the surface of PDMS@Cu‐0 is coated with a relatively dense layer of PDMS, the overall structure is relatively smooth, the surface pores are sparse, and the gas diffusion channels are limited. At a lower voltage (−0.7 V), the PDMS coating begins to loosen its structure, and more medium‐sized pores appear on the surface. The surface roughness of the material increases, which is conducive to contact with reactants. As the voltage increases, the material structure begins to collapse significantly, and a highly cross‐linked 3D porous network is formed on the surface. The number and connectivity of pores are significantly increased, providing more accessible sites and diffusion channels for the electrocatalytic reaction. However, when the voltage is applied to −1.5 V, the catalytic activity decreases instead. We speculate that this may be associated with excessive surface coverage and possible blockage of active sites, which could hinder charge transfer and reactant accessibility [38, 39, 40, 41]. It reveals that PDMS@Cu‐x (x = 0, 1, 2, 3, 4, 5) particles are embedded or partially embedded within the PDMS‐derived amorphous matrix. This morphology indicates that PDMS effectively partially encapsulates the underlying Cu^δ+ species, achieving the dual objectives of our strategy. In other words, establishing localized hydrophilic/hydrophobic domains to regulate CO₂ mass transport pathways and forming the intimate amorphous/crystalline phase interface with Cu^δ+ species to modulate intermediate adsorption while concurrently acting as a protective layer to kinetically stabilize these active phases against excessive reduction.

FIGURE 3.

Morphological and structural characterizations of catalysts. (a) SEM images of PDMS@Cu‐x (x = 0, 1, 2, 3, 4, 5). (b) Pore size distribution of PDMS@Cu‐0 and PDMS@Cu‐4. (c) CV curves of PDMS@Cu‐4. (d) Double‐layer capacitance for different catalysts. (e,f) TOF‐SIMS 3D spatial distribution of K, H, Cu, and O elements for PDMS@Cu‐0 (e) and PDMS@Cu‐4 (f). (g–j) TOF‐SIMS depth profiles of MPCC‐Cu and CC‐Cu. Fragment ions of K (g), H (h), Cu (i), and Cu (j).

N₂ adsorption‐desorption analysis was used to estimate the porosity of PDMS@Cu‐0 and PDMS@Cu‐4 (Figure S10) [42]. Voltage activation changed the pore structure and specific surface area of the PDMS@Cu‐0 material. PDMS@Cu‐4 showed a type‐IV isotherm with a more obvious hysteresis loop than PDMS@Cu‐0, revealing a more obvious mesoporous feature. The detailed pore size distribution is shown in Figure 3b. PDMS@Cu‐4 showed more mesopores in the range of 2–3 nm, explaining the increase in mesopores caused by in situ activation treatment, which is conducive to exposing more active sites. The X‐ray diffraction (XRD) pattern of the PDMS@Cu‐0 was consistent with that of the Cu‐BTC (Figure S11), indicating that the addition of PDMS did not change the crystalline structure of Cu‐BTC. Notably, the peaks of the XRD of the other four catalysts changed significantly compared to the PDMS@Cu‐0 XRD pattern, indicating the existence of catalyst reconfiguration after activation at different voltages (Figure S12).

The electrochemically active surface areas (ECSAs) of different catalysts were compared to detect the intrinsic activity of the catalysts. The cyclic voltammetry (CV) curves obtained at different scan rates to evaluate the double‐layer capacitance (C_dl) for catalysts (Figure 3c,d; Figure S13). It was found that PDMS@Cu‐4 exhibited the largest C_dl value of 1.23 mF cm⁻² (Figure 3e). The ECSA was estimated through electrochemical C_dl measurements in 0.5 m KHCO₃. PDMS@Cu‐4 (20.5 cm² _ECSA) exhibited a higher ECSA compared to PDMS@Cu‐0 (12.7 cm² _ECSA, Figure S14). These results indicate that PDMS@Cu‐4 provides more accessible active sites than PDMS@Cu‐0 and enhances the intrinsic catalytic activity for CO₂RR. To further understand the origin of the enhanced CO₂RR performance, electrochemical kinetics analyses were conducted. As shown in Figure S15, PDMS@Cu‐4 exhibits a significantly smaller semicircle diameter in the Nyquist plot compared with PDMS@Cu‐0, indicating reduced charge‐transfer resistance and accelerated interfacial electron‐transfer kinetics.

Visualization of the 3D distributed catalyst surface by Time‐of‐flight secondary ion mass spectrometry (TOF‐SIMS) allows the spatial distribution of different elements to be visualized. Therefore, the difference in depth composition between PDMS@Cu‐4 and PDMS@Cu‐0 catalysts can be further investigated by TOF‐SIMS (Figure 2e,f; Figures S16 and S17). As shown in Figure 3g,h, it can be found that with the increase of sputtering time, the K element in PDMS@Cu‐4 is higher than that in PDMS@Cu‐0, while the H element is the opposite. This phenomenon reveals a distinct redistribution of K and H species after activation. Although TOF‐SIMS is not strictly quantitative, the consistent trend suggests that the dual‐interface structure may modulate local ion environments, which is correlated with the suppressed HER observed in electrochemical measurements [43]. The change in Cu element is due to the in situ reconstruction of the catalyst after potential activation of PDMS@Cu‐0, which leads to partial dissolution of Cu and reduces Cu content (Figure 3i). The abnormal increase in O content may be due to the large amounts of OH⁻ and CO₃ ²⁻ in the residual electrolyte after voltage activation, which can promote and protect O in Cu₂O (Figure 3j) [44]. This indicates that the catalyst adaptively rebuilds to form the amorphous/crystalline phase interface during the activation process.

2.4. Electronic and Fine Structural Characterizations

XANES/EXAFS spectroscopy has elucidated the chemical state, electronic structure, and coordination environment of Cu species in pristine PDMS@Cu‐4 versus PDMS@Cu‐0. XANES of the Cu K‐edge clearly showed the presence of Cu(I) sites, as evidenced by the observation of a small shoulder peak at around 8983 eV due to 1s → 4p orbital conversion (Figure 4a) [45, 46]. Specifically, the Cu 4p orbitals in Cu(I) sites split into 4p_xy and 4p_z , resulting in two peaks of A₁ and A₂ in the XANES spectra. To clarify the average valence state of Cu species, the Cu K‐edge absorption of various samples was compared with the reference sample, which showed that the average oxidation state of Cu was +0.66 and +1.31 for PDMS@Cu‐4 and PDMS@Cu‐0, respectively (Figure 4b). As shown in Figure 4c, for PDMS@Cu‐4, it exhibits a dominant Cu‐O scattering path at 1.5 Å alongside a characteristic Cu‐O‐Cu interaction at 2.7 Å. This indicates that PDMS@Cu‐4 has a Cu structure similar to Cu₂O. Conversely, PDMS@Cu‐0 displays hybrid signatures, Cu‐O (1.5 Å) coexisting with both Cu‐Cu (1.5 Å) and Cu‐O‐Cu (2.7 Å) bonds, indicating mixed CuO/Cu₂O phase characteristics. Additionally, wavelet‐transformed (WT) EXAFS of the Cu K‐edge oscillations was performed, providing high‐resolution insights in both K and R spaces (Figure 4d), confirming the presence of Cu‐O, Cu‐Cu and Cu‐O‐Cu coordination, consistent with the earlier findings.

FIGURE 4.

Compositions and electronic structures. (a, b) Cu K‐edge XANES (a) and Fourier‐transform k³‐weighted EXAFS (b) spectra for Cu foil, Cu₂O Ref., CuO Ref., PDMS@Cu‐0, and PDMS@Cu‐4. (c) The calculated Cu valence states were derived from XANES spectra. (d) Wavelet‐transformed k³‐weighted EXAFS spectra of PDMS@Cu‐4, PDMS@Cu‐0, CuO Ref., Cu₂O Ref., and Cu foil. (e,f) Cu 2p XPS spectra at different etching times of PDMS@Cu‐0 (e) and PDMS@Cu‐4 (f). (g,h) Cu LMM XPS spectra at different etching times of PDMS@Cu‐0 (g) and PDMS@Cu‐4 (h).

Depth‐resolved XPS (6 nm/min Ar⁺ etching) unveils divergent Cu speciation in PDMS@Cu‐0 and PDMS@Cu‐4 catalysts. Notably, both catalysts exhibited suppressed Cu surface elemental signals. To clarify this phenomenon, we analyzed Si 2p spectra and found that surface Si peaks of PDMS@Cu‐0 and PDMS@Cu‐4 were significantly stronger than those in their interior regions (Figure S18). This suggests that PDMS encapsulation likely suppressed the Cu surface signal. As shown in Figure 4e, Cu²⁺ (935.2 eV) and Cu⁺/Cu⁰ (933.9 eV) states emerge at 36 nm depth [47]. Since it is difficult to distinguish Cu⁺ and Cu⁰ states in the Cu 2p spectra due to the similar binding energies of the two species, X‐ray‐induced Cu LMM Auger electron spectroscopy (XAES) was also employed to identify the surface composition of the samples (Figure 4g). This revealed triphasic speciation at 36 nm (360 s): Cu⁺ (916.9 eV), Cu⁰ (918.4 eV), and Cu²⁺ (917.8 eV) [30, 48, 49]. Conversely, PDMS@Cu‐4 maintains robust Cu⁺ (916.9 eV) and Cu⁰ (918.4 eV) signatures at 36 nm depth (360 s) without detectable Cu²⁺ (Figure 4f,h), confirming the depth‐stable amorphous/crystalline phase interface. This kinetic stabilization of mixed‐valent centres by PDMS confinement aligns with operando XAS findings (Figure 2h,i).

2.5. Electrochemical CO₂RR Performance of Catalysts

The catalytic performance of PDMS@Cu‐0 and PDMS@Cu‐4 for CO₂RR was evaluated in the CO₂‐saturated 0.5 m KHCO₃ electrolyte with an H‐type cell (Figure S19). The gas and liquid products were measured using online gas chromatography (GC) and ¹H nuclear magnetic resonance (NMR) spectroscopy, respectively. The catalytic activity of the two as‐prepared catalysts was first evaluated by linear sweep voltammetry (LSV) in the Ar‐saturated or CO₂‐saturated electrolyte (Figure 5a). PDMS@Cu‐4 exhibits a greater total current density (J) than PDMS@Cu‐0 in CO₂‐saturated electrolyte. This indicates that the catalytic activity of PDMS@Cu‐4 is better than that of PDMS@Cu‐0. To provide a more accurate characterization of the CO₂RR performance of PDMS@Cu‐0 and PDMS@Cu‐4 catalysts, constant‐potential electrolysis was conducted at stepped potentials ranging from −0.7 to −1.5 V. Figure 4b,c shows the FE of the products H₂, HCOOH, CO, CH₄, and C₂H₄ for the two catalysts. Notably, the FE_C2H4 increased from 60.8% for PDMS@Cu‐0 to 86.4% for PDMS@Cu‐4 at the potential of −0.9 V. To better illustrate whether adding PDMS optimized performance, the performance and morphological characterizations of Cu‐BTC were tested (Figure S20). Compared with pristine Cu‐BTC, the PDMS‐modified catalyst exhibits significantly enhanced C₂H₄ selectivity and suppressed HER activity, indicating that PDMS plays an important role in regulating the interfacial microenvironment. Therefore, this improvement correlates with the adaptively reconstructed catalyst surface, which facilitates C‐C coupling kinetics and stabilizes key intermediates.

FIGURE 5.

Electrochemical CO₂RR performance. (a) Total current density of PDMS@Cu‐4 and PDMS@Cu‐0 samples in Ar‐saturated or CO₂‐saturated electrolyte in the voltage range of 0 to −1.83 V at a scanning rate of 0.1 V s⁻¹. (b, c) FEs toward various products for PDMS@Cu‐4 (b) and PDMS@Cu‐0 (c) samples with potentials ranging from −0.7 to −1.5 V in the H‐cell. (d) FE_C2H4 and its corresponding partial current density at −0.9 V for PDMS@Cu‐1 to PDMS@Cu‐5. (e) Ratio of FE_C2H4 to FE_H2 at −0.9 V for PDMS@Cu‐1 to PDMS@Cu‐5. (f) Performance comparison between this work and the reported. (g) The CO₂RR stability test of PDMS@Cu‐4 at −0.9 V in CO₂‐saturated 0.5 m KHCO₃.

To probe the voltage‐dependent activation effects on catalytic behaviour, we systematically tested the electrochemical performance of PDMS@Cu‐0 electrodes subjected to electrochemical preconditioning at varying potentials. The catalytic activity of the catalysts activated at five different voltages in CO₂‐saturated electrolyte was determined by LSV and through ECSA‐normalized activity (Figure S21). Compared with the other four catalysts, the PDMS@Cu‐4 catalyst showed a higher current density and exhibited the best CO₂RR activity. Notably, the FE_C2H4 at −0.9 V exhibited pronounced divergence among the five catalysts, with PDMS@Cu‐4 demonstrating exceptional selectivity (86.4%) toward C₂H₄. This trend was further corroborated by partial current density of ethylene (J_C2H4) measurements, revealing PDMS@Cu‐4 as the optimal candidate for CO₂‐to‐C₂H₄ conversion (Figure 5d).

Comparative analysis of six catalysts at −0.9 V revealed a FE_C2H4/FE_H2 ratio of ∼25 for the optimized PDMS@Cu‐4, surpassing other candidates by ∼2.5 times (Figure 5e; Figure S22). This marked enhancement in FE_C2H4/FE_H2 ratio underscores its intrinsic activity toward CO₂‐to‐C₂H₄ conversion, likely attributed to the dual interface formed after activation reconstruction stabilized ^*CO intermediates while suppressing hydrogen evolution. We compared the CO₂RR performance of our designed PDMS@Cu‐4 catalyst with previously reported FE_C2H4, J_C2H4, and voltage, demonstrating the advanced nature of our designed catalyst for high‐performance CO₂ electrolysis (Figure 4f; Supplementary Table S1). Finally, PDMS@Cu‐4 was continuously electrolyzed at −0.9 V for 24 h, and its FE_C2H4 remained ∼80% (Figure 4g). Simultaneously, the degradation of J_C2H4 was negligible, demonstrating excellent durability for continuous CO₂‐to‐C₂H₄ conversion. The electronic structures (Figure S23) and morphologies (Figures S24 and S25) remain almost the same as those of the catalyst before long‐term electrolysis.

To further evaluate the catalyst performance under enhanced CO₂ mass‐transport conditions, a membrane electrode assembly (MEA) system was employed (Figure S26a). As shown in Figure S26, PDMS@Cu‐4 maintains high C₂H₄ selectivity at increased current densities compared to the H‐cell. Specifically, FE_C2H4 reached 65% at a total current density of 150 mA cm⁻². The performance under the MEA system further supports the role of the reconstructed dual interface in facilitating reactant transport and maintaining efficient C‐C coupling.

3. Conclusion

In this work, we propose an in‐situ activation dual strategy to construct localized dual interfaces that decouple mass‐charge transfer. This design modulates the electronic structure of catalytic active sites, optimizing the adsorption of key intermediates and facilitating charge transfer, thereby enhancing ethylene selectivity. The interfacial regulation is further corroborated by in‐situ characterizations and DFT calculations. Remarkably, the dual‐interface engineered PDMS@Cu‐4 catalyst achieves an impressive FE_C2H4 of 86% at low overpotential with a FE_C2H4/FE_H2 ratio of ∼25, which is approximately 2.5 times higher than that of other reported catalysts while maintaining operational stability for 24 h. This decoupling‐mediated strategy thus simultaneously optimizes CO₂ mass transport and intermediate adsorption kinetics, enabling highly efficient ethylene electrosynthesis.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma73692‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.