All chemical reagents were used as received without further purification. Specifically, cobalt (II) nitrate hexahydrate (Co(NO₃)₂·6H₂O, 99%), zinc chloride (ZnCl₂), tetraethyl orthosilicate (TEOS, > 99%), urea (99.5%), ammonium hydroxide solution (25%~28%), and cetyltrimethylammonium bromide (CTAB) were purchased from Sinopharm Chemical Reagent Co., Ltd. Deionized water was employed for the preparation of all solutions throughout the experimental process.

Scanning electron microscopy (SEM) was conducted on a Zeiss GeminiSEM300 instrument (Carl Zeiss Microscopy Ltd.). Transmission electron microscopy (TEM) and high-angle annular dark-field scanning TEM (HAADF-STEM) with energy-dispersive X-ray spectroscopy mapping were performed using a FEI Talos F200X instrument (Thermo Fisher Scientific, Waltham, MA, USA). X-ray photoelectron spectroscopy (XPS) was carried out on an ESCALAB 250Xi instrument (Thermo Fisher Scientific, Waltham, MA, USA) using monochromatic Al Kα radiation. X-ray diffraction (XRD) patterns were collected on a Rigaku Ultimate IV diffractometer (Rigaku Corporation, Tokyo, Japan) with Cu Kα radiation. Nitrogen adsorption-desorption isotherms were measured using a Micromeritics 3Flex instrument (Micromeritics Instrument Corporation, Norcross, GA, USA). For in situ attenuated total reflection-fourier-transform infrared spectroscopy (ATR-FTIR) studies, a Thermo Fisher Nicolet IS 50 FTIR spectrometer configured with an MCT/B detector and a VeeMAX III ATR accessory was utilized. A custom three-electrode spectroelectrochemical cell was employed. The working electrode was fabricated by applying a catalyst film onto a gold-sputtered silicon prism. An Ag/AgCl electrode and a platinum wire served as the reference and counter electrodes, respectively. The electrochemical environments consisted of two separate electrolyte systems: (i) 0.1 M KHCO₃ aqueous solution saturated with CO₂, and (ii) the hybrid electrolyte system composed of 0.06 M KHCO₃ + 0.1 M [EMIM]Cl saturated with CO₂. The cell was interfaced with a potentiostat, and IR spectra were collected during controlled potential holds across a range from open-circuit potential, 0 to -1.1 V vs. reversible hydrogen electrode (RHE).

The electrochemical measurements were performed using a CHI760E electrochemical workstation (CH Instruments, Shanghai) in a conventional two-compartment H-cell separated by a Nafion 117 membrane. A standard three-electrode configuration was employed, consisting of a catalyst-coated working electrode, an Ag/AgCl reference electrode, and a Pt sheet counter electrode. Electrochemical characterization involved linear sweep voltammetry (LSV) conducted at a scan rate of 5 mV/s in CO₂-saturated and Ar-saturated electrolytes. The hybrid electrolyte system was prepared by mixing 0.1 M KHCO₃ aqueous solution with 1-ethyl-3-methylimidazolium chloride ([EMIM]Cl) in a 3:2 volume ratio. (The hybrid electrolyte system was prepared by first diluting commercial [EMIM]Cl (≥ 98% purity, molecular weight 146.62 g/mol) with deionized water to prepare a 0.25 M [EMIM]Cl stock solution. e.g., 30 mL of 0.1 M KHCO₃+20 mL of [EMIM]Cl for larger-scale preparations, or 15 mL + 10 mL for smaller-scale experiments, 0.06 M KHCO₃ + 0.1 M [EMIM]Cl. The final electrolyte thus contains approximately 0.06 M KHCO₃ and 0.1 M [EMIM]Cl. Electrochemical impedance spectroscopy (EIS) measurements were carried out over a frequency range of 100 kHz to 0.1 Hz with an AC amplitude of 5 mV. The electrocatalytic CO₂RRs were conducted under potentiostatic control in both a 0.1 M KHCO₃ solution and a hybrid electrolyte system comprising 0.06 M KHCO₃ + 0.1 M [EMIM]Cl. A constant CO₂ flow of 30 mL/min was maintained throughout the electrolysis experiments, with each compartment of the H-cell containing 25 mL of electrolyte. The gaseous effluent from the cathodic compartment was continuously monitored and quantitatively analyzed using online gas chromatography. The FE for a given product i was determined by the relation: FE_i = (n_i × z_i × F)/Q × 100%, where the terms n_i, z_i, F, and Q denote the moles of product generated, the electron transfer number per molecule, the Faraday constant, and the total charge consumed, respectively. Potentials referenced to the Ag/AgCl electrode were converted to the RHE scale using the formula: E (vs. RHE) = E (vs. Ag/AgCl) + 0.197 + 0.0596 × pH (The pH measurements were performed at 27 °C, and the Nernstian slope was calculated accordingly as 0.0596 V per pH unit). First-principles calculations based on DFT were carried out using the Vienna Ab-initio Simulation Package (VASP). The projector augmented-wave (PAW) method and the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional were employed in the calculations. DFT computations were performed to systematically investigate the interaction energies between the [EMIM]⁺ cation and CO₂ molecules, as well as reaction intermediates (COOH and OCHO) on Co and Zn sites. The reaction energy barriers in different systems were also evaluated.

The CoZn oxide catalysts supported on MSN were first examined by SEM and TEM. Regardless of the Co:Zn ratio, all catalysts self-assemble into uniform three-dimensional (3D) flower-like architectures composed of interlaced nanosheets (SEM images of the catalysts are provided in Supplementary Figure 1). This open hierarchical structure provides a high density of accessible active sites and facilitates reactant diffusion, a distinct structural advantage for CO₂RR^[21,22]. Increasing the Zn content systematically reduces the nanoflower size [Figure 1A-C]. Mechanistically, Zn²⁺ ions promote nucleation while suppressing Ostwald ripening, acting as a “morphology-directing agent” that balances nucleation promotion and growth inhibition^[23,24]. Elemental mapping [Figure 1D] confirms homogeneous distribution of Co, Zn, and Si, verifying successful catalyst loading.

The crystal phase composition and structural evolution of the mesoporous SiO₂-supported CoZn oxide catalysts were probed by XRD. As depicted in Figure 2A, the bare mesoporous SiO₂ (SiO₂-supported CoZn oxide catalysts) support exhibits a characteristic broad diffraction halo centered at around 22.47°, which is indicative of its amorphous nature. Upon the integration of Co and Zn species, distinct crystalline phases emerge. For the catalysts with Co:Zn ratios of 2:1 and 1:1, the XRD patterns reveal well-defined diffraction peaks at 35.31° and 58.90°^[25,26]. These peaks can be indexed to the characteristic eflections of a cobalt-zinc composite oxide phase, suggestive of the formation of a crystalline spinel-type structure (e.g., ZnCo₂O₄)^[27]. Notably, a significant phase evolution is observed with a further increase in Zn content. Specifically, the pattern for the Co:Zn = 1:2 catalyst not only retains the aforementioned peaks but also manifests an additional distinct peak at 64.64°. The emergence of this new diffraction feature coincides with the overall pattern more closely aligning with that of a zinc-rich phase, such as ZnO (zincite, JCPDS No. 36-1451), where the peak at ~64.64° corresponds to its (103) crystal plane^[28]. This systematic shift in the XRD profiles unequivocally demonstrates that increasing Zn incorporation redirects the crystallization pathway, favoring the formation of ZnO alongside or in place of the bimetallic spinel phase, thereby altering the catalyst's intrinsic phase composition. It should be noted that the as-synthesized catalysts are oxide-based materials (ZnCo₂O₄ spinel and ZnO phases), as confirmed by XRD analysis, rather than CoZn metallic alloys. CoZn oxide nanoflower catalysts with different Co:Zn atomic ratios exhibit distinct pore structural characteristics, which are closely correlated with their CO₂RR catalytic performance and product selectivity. From the pore size distribution curves [Figure 2B], clear composition-dependent mesoporous features can be observed across the three catalysts. The Co:Zn = 2:1 catalyst shows a dominant mesopore distribution in the 10-20 nm range, with a differential pore volume of ~0.4 cm³/g. This narrower and more uniform mesoporous structure effectively restricts the over-hydrogenation of intermediates and promotes CO desorption, leading to a FE of 80.4% for CO at this ratio. In contrast, for the Co:Zn = 1:2 catalyst, a sharp and intense mesopore peak centered at approximately 10-40 nm is evident, with a differential pore volume (dV/dlog Pore Volume) reaching ~0.6 cm³/g, indicating the formation of a highly developed and uniform mesoporous network. This large-sized mesoporous structure facilitates the rapid diffusion of CO₂ and reaction intermediates, providing sufficient reaction space for the formation of HCOOH, thereby achieving a FE of 70.9% for formic acid at this ratio. The Co:Zn = 1:1 catalyst, however, shows a relatively broad pore-size distribution with a distinct peak at smaller pore sizes, and delivers a lower maximum differential pore volume. This observation aligns well with its moderate catalytic activity and broad product distribution trend in CO₂RR.

To establish a comprehensive correlation between catalyst composition, physicochemical properties, and CO₂RR performance, we first conducted systematic compositional and textural characterization of the three CoZn oxide nanoflower catalysts. Inductively coupled plasma optical emission spectroscopy (ICP-OES) analysis [Supplementary Table 1] confirms that the actual Co:Zn molar ratios (0.98:1, 1.95:1, and 0.49:1 for the three catalysts, respectively) are in excellent agreement with the nominal target ratios (≈1:1, ≈2:1, and ≈1:2), validating our precise synthetic control over catalyst composition. Complementary Brunauer-Emmett-Teller (BET) surface area and total pore volume measurements [Supplementary Table 2] reveal systematic textural variations as a function of composition. The Co:Zn = 2:1 catalyst exhibits a BET surface area of 142 m²/g and a total pore volume of 0.38 cm³/g (with a pore size of approximately 20 nm), whereas the Co:Zn = 1:2 catalyst shows substantially higher values of 186 m²/g and 0.57 cm³/g, consistent with its larger mesopores centered at roughly 25-40 nm [Figure 2B]. The Co:Zn = 1:1 catalyst displays intermediate values (165 m²/g and 0.45 cm³/g). These systematic textural variations directly correlate with the observed product selectivity: the narrower mesopores of the Co-rich catalyst favor CO desorption and suppress over-hydrogenation, while the larger mesopores of the Zn-rich catalyst facilitate rapid diffusion of CO₂ and reaction intermediates, providing sufficient reaction space for HCOOH formation. Moreover, XPS survey scans [Supplementary Table 3] reveal that the residual nitrogen content from the CTAB template and urea is consistently low (~1.2-1.5 at%) across all three catalysts with no systematic variation, confirming that residual nitrogen does not contribute to the observed selectivity trends. Having established the composition-property relationships, we next examined the electronic structure modulation induced by the Co:Zn ratio using high-resolution XPS analysis. XPS was employed to gain deep insights into the surface elemental chemical states and electronic structures of the mesoporous SiO₂ and the CoZn oxide nanoflower catalysts supported thereon. High-resolution scans of the C 1s region revealed characteristic peaks at 283.0, 284.1, 285.4 and 287.3 eV [Figure 2C], which are assigned to metal carbide/carbon-silicon bonds (C-M/C-Si), adventitious carbon/amorphous carbon (C-C/C-H, commonly used as an internal reference), hydroxyl/ether carbon (C-O) and carboxyl/carbonyl carbon (C=O), respectively^[29]. Notably, a negative binding energy shift of 0.1-0.4 eV was observed for the C-O and C=O peaks with decreasing Co:Zn ratio (i.e., increasing Zn content), indicating enhanced electron-donating character in Zn-rich regions and a weakening of the C-O bond polarity^[30,31]. For the Si 2p spectrum [Figure 2D], the pristine mesoporous SiO₂ support exhibited peaks at 101.5, 101.8, and 102.5 eV, corresponding to Si²⁺ (sub-oxide), Si³⁺ (intermediate oxide), and Si⁴⁺ (SiO₂), respectively, confirming its partially oxidized surface state^[32,33]. However, upon deposition of the CoZn oxide nanoflowers, the Si 2p peaks underwent a systematic negative shift (up to 0.8-0.9 eV) to 100.6, 101.6, and 102.2 eV. This clearly demonstrates electron transfer from the metal nanoflowers to the SiO₂ support, reducing the silicon oxidation state, with the effect being more pronounced at higher Zn contents due to promoted interfacial electron coupling and delocalization facilitated by Zn's electronegativity. Further analysis of the metal active sites showed that the Co 2p spectrum featured main peaks at 780.2 eV (Co 2p_3/2) and 796.5 eV (Co 2p_1/2) [Figure 2E], indicative of Co²⁺ oxidation states (in phases like CoO or ZnCo₂O₄ spinel), along with satellite peaks at 785.3 and 802.4 eV consistent with a high-spin d⁷ electronic configuration^[34,35]. As the Co:Zn ratio decreased, the Co 2p_3/2 main peak exhibited a positive binding energy shift (up to +0.5 eV), revealing a decreased electron density at the Co sites, originating from the competition of Zn²⁺ for lattice oxygen, which weakens the covalency of the Co-O bond^[36]. In stark contrast, the Zn 2p spectrum [Figure 2F], with characteristic peaks at 1,021.0 eV (Zn 2p_3/2) and 1,043.9 eV (Zn 2p_1/2), confirming the presence of Zn²⁺ species, showed a negative shift of approximately 0.4 eV for the Zn 2p_3/2 peak when the Co:Zn ratio increased^[37]. This suggests an increased electron density at Zn sites, likely associated with electron transfer from Co²⁺ to the Zn 4s-p orbitals, thereby optimizing the interfacial Lewis acidity. XPS analysis unequivocally reveals that variations in the Co:Zn ratio modulate the intermetallic charge distribution, inducing systematic binding energy shifts (positive for Co, negative for Zn and SiO₂). This pronounced electronic effect profoundly influences the adsorption behavior of key reaction intermediates on the catalyst surface, thereby providing crucial evidence at the electronic structure level for understanding the switching mechanism of the CO₂RR pathway selectivity (CO vs. HCOOH).

EIS analysis reveals significant differences between the conventional KHCO₃ electrolyte and the innovative [EMIM]Cl/KHCO₃ composite electrolyte in the electrocatalytic CO₂RR. Bode plots demonstrate that the impedance modulus of the composite electrolyte system is markedly lower than that of the pure KHCO₃ system at the same applied potential, indicating a substantial reduction in interfacial charge transfer resistance (Nyquist plots, Figure 3A and B). The difference is most pronounced at less negative potentials and diminishes at more negative potentials due to the increased driving force for charge transfer, which accelerates the kinetics in both systems. Quantitative analysis using equivalent circuit fitting [Supplementary Table 4] confirms that the charge transfer resistance (Rct) of the hybrid electrolyte is consistently lower than that of pure KHCO₃ across all tested potentials. This phenomenon is attributed to the unique interfacial layer formed by the adsorption of [EMIM]⁺ on the electrode surface, which interacts with CO₂ molecules to form [EMIM]⁺-CO₂ complex intermediates, substantially lowering the overpotential for CO₂ reduction to CO^[14,38,39]. As the potential shifts negatively from -0.3 to -0.9 V vs. RHE, the impedance values of both electrolytes gradually decrease, reflecting enhanced charge transfer kinetics at higher overpotentials. However, the composite electrolyte consistently exhibits lower impedance values across all tested potentials, demonstrating the additional catalytic promotion provided by the imidazolium cations^[38]. The evolution of phase and phase angle plots offers further critical insights into the interfacial kinetics (Bode magnitude plots, Figure 3C and D). The shift of the phase angle peak to higher frequencies in the composite electrolyte indicates a decrease in the time constant of the electrode interface process, corresponding to an accelerated charge transfer rate (Bode phase-angle plots, Figure 3E and F). The sharpening of the phase angle peak suggests a more concentrated distribution of time constants, reflecting improved uniformity of the electrode/electrolyte interface and enhanced synergy in the reaction process. The overall shift of the phase angle peak to more negative values (e.g., from -30° to -40°) signifies enhanced interfacial capacitive behavior, originating from the restructured electric double layer at the electrode surface with higher double-layer capacitance due to the imidazolium cations. This reconstituted double layer not only promotes CO₂ enrichment but also reduces the reaction activation barrier by stabilizing the CO₂ radical anion intermediate. The interfacial enhancement mechanism in the composite electrolyte system can be summarized in three aspects: First, the [EMIM]⁺ adsorbed on the electrode surface forms hydrogen bonds with the CO₂ radical anion via C₂-H or C₄/C₅-H, stabilizing key reaction intermediates^[40]. Second, the adsorption layer of [EMIM]⁺ alters the structure and thickness of the electric double layer^[41], increasing the local electric field strength and facilitating the initial electron transfer step to CO₂^[42]. Finally, this unique interfacial microenvironment may also suppress the competing HER, further improving the selectivity of CO₂RR. These factors collectively contribute to the substantially enhanced efficiency of CO₂RR, providing an important theoretical foundation for the design of highly efficient electrocatalytic interfaces^[43]. It is worth emphasizing that the advantages of our [EMIM]Cl/KHCO₃ hybrid electrolyte system are fundamentally distinct from and complementary to those provided by gas diffusion electrode (GDE) architectures. Our hybrid electrolyte actively participates in the reaction mechanism through multiple functions that GDE architecture alone cannot provide: (i) the [EMIM]⁺ cations form strong non-covalent interactions with the negatively charged CO₂^- intermediate via C₂-H hydrogen bonding, stabilizing this high-energy species and lowering the activation barrier for the initial electron transfer step; (ii) the electrolyte establishes a linear hydrogen-bonded [EMIM]-H···O-H···O-C(O)^- pre-reaction complex that enables a cooperative proton tunneling mechanism, fundamentally altering the PCET kinetics and substantially reducing the overpotential; and (iii) the imidazolium cation adsorption layer inherently suppresses HER by excluding water molecules from the active sites. These are intrinsic molecular-level benefits that arise from the electrolyte's chemical composition, independent of the electrode architecture. While GDEs effectively address the physical mass transport limitations of CO₂, our electrolyte engineering strategy targets the thermodynamic and kinetic bottlenecks at the molecular level, offering a synergistic and complementary design principle that can be readily integrated with GDE-based systems for enhanced practical performance.

This study systematically investigates the regulatory mechanism and synergistic enhancement effects of CoZn oxide nanoflower catalysts in a [EMIM]Cl/KHCO₃ composite electrolyte system for the electrocatalytic CO₂RR. In the conventional KHCO₃ electrolyte [Figure 4A], the LSV curves of all CoZn oxide catalysts under CO₂ and Ar atmospheres are nearly superimposable, indicative of negligible electrocatalytic activity toward CO₂RR. Here, the current response for CO₂RR is indistinguishable from the background current associated with the HER, highlighting the inefficiency of bare KHCO₃ in facilitating CO₂ activation. In stark contrast, incorporation of the [EMIM]Cl/KHCO₃ electrolyte [Figure 4B] elicits a dramatic enhancement in cathodic current density for all catalysts under CO₂, with the Co:Zn = 2:1 catalyst exhibiting a significantly enhanced current density (P < 0.01, Student's t-test, n =3) compared to the other compositions. This observation unequivocally demonstrates that the composite electrolyte synergizes with the CoZn oxide catalysts to accelerate CO₂RR kinetics and suppress parasitic side reactions. In the pure KHCO₃ electrolyte [Figure 4C-E], the product landscape is dominated by CO, HCOOH, and H₂, with substantial HER contribution across all potentials and catalyst compositions. A clear composition-dependent selectivity trend emerges: increasing Zn content shifts the primary product from CO (Co:Zn = 2:1) to HCOOH (Co:Zn = 1:2), while the Co:Zn = 1:1 catalyst yields a mixed distribution of CO and HCOOH [Table 2]. This selectivity divergence stems from intrinsic differences in intermediate binding energies at Co and Zn active sites: Co sites bind *COOH and *CO intermediates with moderate strength, favoring the sequential proton-electron transfer steps leading to CO desorption. Conversely, Zn sites exhibit negligible affinity for *CO and preferentially stabilize the *OCHO intermediate, thus steering the reaction toward HCOOH formation via the formate pathway. The product distribution is profoundly altered in the [EMIM]Cl/KHCO₃ electrolyte [Figure 4F-H]. Most notably, the Co:Zn = 2:1 catalyst achieves a total FE of 89.5% for C₁ products (CO + HCOOH) at a low overpotential of -0.9 V vs. RHE in the [EMIM]Cl/KHCO₃ hybrid electrolyte [Figure 4F]. The maximum total C₁ FE of 93.81% ± 1.8% is achieved at -1.0 V vs. RHE, accompanied by a marked suppression of HER^[7,44-47]. This performance enhancement is attributed to the multi-faceted synergistic effects of the ionic liquid component: (1) [EMIM]⁺ cations form a robust interfacial adsorption layer, where their C₂-H moieties engage in strong non-covalent interactions with negatively charged *CO₂^- intermediates, stabilizing this high-energy species and lowering the activation barrier for CO₂ reduction; (2) the high CO₂ solubility and viscosity of the ionic liquid establish a CO₂-enriched microenvironment at the electrode electrolyte interface, ensuring a sustained supply of reactant; (3) the unique electric double-layer structure induced by [EMIM]Cl generates a strong interfacial electric field, which polarizes and activates CO₂ molecules, while simultaneously excluding water molecules from the active sites to kinetically suppress HER. Collectively, these effects enable the precise compositional tuning of Co/Zn ratios to act as a "selectivity switch" for CO₂RR pathways, underscoring the critical role of electrolyte-catalyst engineering in achieving efficient and selective CO₂ conversion. To evaluate the stability of [EMIM]Cl under our operating conditions, we performed ¹H NMR (nuclear magnetic resonance) spectroscopy on the [EMIM]Cl/KHCO₃ hybrid electrolyte before and after 12 h of chronoamperometry at -1.0 V vs. RHE, which corresponds to the optimal CO₂RR potential for the Co:Zn = 2:1 catalyst. The NMR spectra (presented in Supplementary Figure 2) show that all characteristic proton resonances of the [EMIM]⁺ cation, specifically, the C₂-H proton at ~9.2 ppm, the C_4/5-H protons at ~7.5 ppm, the N-CH₂ protons at ~4.2 ppm, and the N-CH₃ protons at ~3.8 ppm, remain essentially unchanged in both chemical shift and integration intensity after prolonged electrolysis. Importantly, no new peaks corresponding to potential decomposition products (such as deprotonated imidazole species or ring-opened products) were detected within the detection limit of NMR spectroscopy. Additionally, post-electrolysis FTIR analysis of the electrolyte showed no emergence of new absorption bands that would indicate degradation of the imidazolium cation. To investigate the structural evolution of the oxide-based CoZn catalysts under CO₂RR operating conditions, we performed post-electrolysis characterization of the Co:Zn = 2:1 catalyst after 12 h of chronoamperometry at -1.0 V vs. RHE in the [EMIM]Cl/KHCO₃ electrolyte. As shown in Supplementary Figure 3, the XRD pattern of the spent catalyst retains the characteristic diffraction peaks of the spinel ZnCo₂O₄ phase, although a slight decrease in peak intensity is observed, suggesting that the overall crystalline framework remains largely intact. Post-electrolysis high-resolution XPS analysis of the spent Co:Zn = 2:1 catalyst [Supplementary Figure 4A-C] reveals notable yet subtle changes in the Co 2p spectrum compared to the pristine catalyst. Specifically, a small shoulder emerges at approximately 778.5 eV, which is characteristic of metallic Co⁰ formed via the electrochemical reduction of Co²⁺ species under cathodic polarization (Co²⁺ + 2e^- → Co⁰). This observation confirms that a minor fraction of surface Co²⁺ species undergoes reduction during CO₂RR electrolysis. Concurrently, the main Co 2p_3/2 peak remains centered at approximately 780.2 eV, indicative of Co²⁺ species in the spinel oxide structure, while a small feature at approximately 781.5 eV, previously ambiguously assigned, is more appropriately attributed to Co³⁺ species inherently present in the octahedral sites of the ZnCo₂O₄ spinel lattice or to surface hydroxylation resulting from air exposure during sample transfer, rather than to cathodic reduction. The Zn 2p spectrum, in contrast, exhibits no discernible changes after electrolysis. Collectively, these observations indicate that the majority of Co species retain the Co²⁺ oxidation state, confirming that the spinel oxide framework is largely preserved under reaction conditions. The emergence of the metallic Co⁰ shoulder, however, suggests that only a minor fraction of surface Co²⁺ species undergoes reduction during electrolysis, potentially generating the true active sites for CO₂RR while the bulk spinel structure remains intact. Importantly, the characteristic spinel oxide phase is maintained as the dominant structure, with only minor surface reconstruction occurring at the outermost atomic layers. These results collectively demonstrate that the CoZn oxide nanoflowers serve as stable precatalysts that undergo minimal structural evolution while preserving their hierarchical architecture, with surface reduction potentially generating the true active sites for CO₂RR.

In situ Fourier transform infrared spectroscopy was employed to systematically compare the mechanistic pathways of CO₂RR over CoZn oxide catalysts in conventional KHCO₃ vs. a [EMIM]Cl/KHCO₃ electrolytes. In the conventional KHCO₃ electrolyte [Figure 5A and B], the characteristic spectral signals corresponding to the C-OH deformation and asymmetric O-C-O stretching vibrations of the *COOH intermediate (within the 1,190-1,350 cm^-1 region) and chemisorbed *CO (at 1,015 cm^-1) remained weak, showing only a gradual increase as the potential shifted negatively. This indicates limited catalytic activity toward the CO formation pathway under these conditions. In contrast, within the [EMIM]Cl/KHCO₃ electrolyte [Figure 5C and D], the signal intensities in the same spectral regions were substantially enhanced and rose rapidly with increasingly negative potential. This observation can be attributed to the ionic liquid's role in elevating the local CO₂ concentration and modulating the interfacial electric field, thereby effectively promoting the formation of the *COOH intermediate and the desorption of *CO. Simultaneously, the Zn-C vibrational peak observed around ~842 cm^-1 reveals the synergistic role of Zn sites, whose electronic interaction with the ionic liquid may further optimize the electronic structure of adjacent Co active centers.

To systematically elucidate the intrinsic influence of the electrolyte microenvironment on the performance of the electrocatalytic CO₂RR, this study integrates macroscopic electrochemical experiments with multiscale theoretical simulations. By constructing three characteristic interfacial models, namely, the conventional KHCO₃ aqueous solution, the anhydrous [EMIM]Cl ionic liquid, and [EMIM]Cl/KHCO₃, and combining DFT calculations with ab initio molecular dynamics (AIMD) simulations, we unravel at the molecular level the regulatory mechanism of the interfacial hydrogen-bonding network on proton transfer pathways and reaction energy barriers. The computational results reveal that in the conventional KHCO₃ system, the interface features a disordered weak interaction network of K⁺···OH^-···OC(O)^- formed by water molecules and K⁺, with a binding energy of only -1.1 eV [Figure 6A]. This nonspecific adsorption is identified as the key factor responsible for the high overpotential. In the pure [EMIM]Cl system, although [EMIM]⁺ cations are enriched at the cathode interface, proton transfer is severely hindered due to steric hindrance between the imidazole ring and the ·CO₂^- intermediate, as well as the excessively high energy barrier for direct cleavage of the C₂-H bond. The binding energy is even lower (-0.73 eV, Figure 6B), confirming the thermodynamic and kinetic unfavorability of the direct proton transfer pathway in the absence of a proton carrier. Upon introducing water to form the [EMIM]Cl/KHCO₃ hybrid electrolyte, the interfacial proton transfer mechanism undergoes a fundamental transformation. DFT calculations reveal that water molecules act as proton-bridging media, simultaneously forming strong hydrogen bonds with the C₂-H of [EMIM]⁺ and the oxygen atom of ·CO₂^-, thereby constructing a linear [EMIM]-H···O-H···O-C(O)^- pre-reaction complex. This configuration, through cooperative hydrogen bonding, substantially enhances the binding energy to -3.02 eV [Figure 6C], greatly stabilizing the key intermediate. The core of this mechanism lies in the synergistic proton tunneling process: the formation of the hydrogen-bond chain weakens the C₂-H and O-H bonds, coupling proton transfer with electron donation. This process achieves functional separation [EMIM]⁺ serves as the proton donor, while water molecules act as an efficient proton transfer shuttle, thereby substantially reducing the activation energy of the PCET step.

Furthermore, the adsorption behavior of the pre-reaction complex [EMIM]-H···O-H··· on the CoZn oxide nanoflower catalyst surface was calculated, yielding a binding energy as high as -1.47 eV [Figure 6D], confirming the effective stabilization of this active species on the catalyst surface. Combined with the analysis of the reaction free energy diagram [Figure 6E], the thermodynamic favorability of this pathway for CO production is validated. To capture the dynamic interfacial behavior under realistic operating conditions, we performed AIMD simulations [Figure 6F]. The results show that during a 10 ps dynamic evolution, water molecules continuously construct and reconstruct transient linear [EMIM]-H···O-H···O-C(O)^- hydrogen-bond networks between [EMIM]⁺ and ·CO₂^-. This dynamic hydrogen-bridging mechanism provides atomistic evidence for the existence and sustainability of the aforementioned synergistic proton tunneling pathway, revealing the efficient synergy between proton supply and transfer channels in the hybrid electrolyte system. These findings offer a clear molecular-level foundation for designing low-overpotential electrolyte environments for CO₂RR.