Ultrastable Cu‐Based Dual‐Channel Heterowire for the Switchable Electro‐/Photocatalytic Reduction of CO2

Abstract Catalytic conversion of CO2 into high value‐added chemicals using renewable energy is an attractive strategy for the management of CO2. However, achieving both efficiency and product selectivity remains a great challenge. Herein, a brand‐new family of 1D dual‐channel heterowires, Cu NWs@MOFs are constructed by coating metal–organic frameworks (MOFs) on Cu nanowires (Cu NWs) for electro‐/photocatalytic CO2 reductions, where Cu NWs act as an electron channel to directionally transmit electrons, and the MOF cover acts as a molecule/photon channel to control the products and/or undertake photoelectric conversion. Through changing the type of MOF cover, the 1D heterowire is switched between electrocatalyst and photocatalyst for the reduction of CO2 with excellent selectivity, adjustable products, and the highest stability among the Cu‐based CO2RR catalysts, which leads to heterometallic MOF covered 1D composite, and especially the first 1D/1D‐type Mott–Schottky heterojunction. Considering the diversity of MOF materials, the ultrastable heterowires offer a highly promising and feasible solution for CO2 reduction.


Introduction
Carbon dioxide (CO 2 ), the major factor in global warming, is an abundant and inexpensive C1 building block to provide high value-added chemicals and/or fuels. [1,2]Accordingly, efficient catalytic conversion of CO 2 has great significance for the sustainable development of human society, green synthesis of fine chemicals and clean energy for carbon neutrality. [3,4]Among the diverse DOI: 10.1002/advs.202302881[34][35][36] However, the relatively low stability and the uncontrolled competing hydrogen evolution reaction (HER) limit the selectivity of target products and hinder further applications of Cu NWs. [37][43][44][45] Herein, we used Cu NWs as the "electron channel" to conduct, transmit and directionally supply electrons for improving the current efficiency, and rationally grew a layer of MOFs on the surface of Cu NWs as the "molecule channel" to auto-concentrate CO 2 molecules, control the entrance of H 2 O molecules, and switch the molecular productions for optimizing the stability, conversion rate and the selectivity.Consequently, a brand-new family of heterowires, Cu NW@MOFs, were constructed, which acted as switchable 1D dual-channel catalysts with superior stability to accurately manipulate the CO 2 RR production from relatively pure CO to complete syngas with broadly variable CO/H 2 ratios from 1:1 to 1:3.2.To further illustrate the role of these two channels in the heterowire catalysts, a www.advancedscience.comphotocatalytic heterowire, Cu NWs@HKUST-1, was constructed by using the semiconducting MOF cover as both molecule channel and photoelectric converter, which exhibited the first 1D/1D-type Mott-Schottky heterojunction that significantly boosted the activity for photocatalytic CO 2 reduction to CO, confirming the molecule and electron communication between the dual channels of the heterowire electro-/photocatalysts.This contribution provides a reliable solution and materials for both CO 2 management and green syngas-supply toward vast synthetic requirements with all-round superiorities including stability, efficiency, and product economy.

Construction of Cu NWs@ZIF-8 Heterowires
Cu NWs were hydrothermally synthesized according to reported approaches. [46,47]Then, a layer of ZIF-8 was in-situ wrapped on the surface of the Cu NWs with hexadecyltrimethylammonium bromide (CTAB) [48][49][50] as surfactants (Figure 1a).[53] To coat MOF ZIF-8 onto Cu NWs, the fresh Cu NWs were first dispersed in degassed methanol, and dual-role CTAB was added as both a surfactant and protectant with further ultrasonic dispersion.Then, solutions of dimethylimidazole ligand and zinc nitrate in degassed methanol were sequentially added to the above dispersion under N 2 .Finally, a light pink sample of Cu NWs@ZIF-8 was obtained after 1 h of reaction by centrifugation under the protection of an inert gas and was dried under dynamic vacuum.
Structure and morphology of the as-prepared Cu NWs and Cu NWs@ZIF-8 were characterized by powder X-ray diffraction (PXRD), synchrotron PXRD, scanning electron microscopy (SEM) and transmission electron microscopy (TEM).The diffraction pattern (Figure S1, Supporting Information) of the asprepared red-brown Cu NWs shows three sharp peaks with high diffraction intensities at 43.3°, 50.4°, and 74.1°, corresponding to (111), (200), and (220) of Cu (PDF#04-0836).The SEM and low-resolution TEM images confirm the uniform nanowire structure of Cu NWs with a diameter of 22.1±3.5 nm and a length of ≈10 μm (Figure 1b,c; Figures S2 and S3, Supporting Information).The X-ray photoelectron spectroscopy (XPS) spectrum (Figure S4a,b, Supporting Information) of the as-prepared Cu NWs shows two peaks at binding energies of 932.6 and 952.5 eV, which could be attributed to Cu2p in Cu/Cu 2 O.In addition, there are peaks at binding energies of 934.4 and 954.4 eV accompanied by satellite peaks at ≈941.9, 944.1, and 962.5 eV, which are assigned to CuO, [54] indicating that the surface of the Cu NWs is easily oxidized in air.Nevertheless, this oxidized surface of Cu NWs is easily recovered to Cu under reaction conditions through both XPS of a sample etched at a depth of 2 nm and CV activation within 50 cycles (Figure S4c,d, Supporting Information).
To characterize the fine structure of the Cu NWs@ZIF-8 composite, synchrotron PXRD ( = 0.77471 Å) was used to analyze the phase structure.Benefiting from the high resolution of synchrotron PXRD (Figure 1g), characteristic diffraction peaks of ZIF-8 were clearly observed in the range of 4°-20°, the intense Cu NWs peaks were maintained, and no other side peaks were observed, indicating the successful combination of Cu NWs with ZIF-8 and the well-preserved crystalline phases of both Cu NWs and the MOF cover.Infrared spectroscopy confirmed that there was no CTAB residue in the Cu NWs@ZIF-8 composite (Figure S5, Supporting Information).X-ray absorption fine structure (XAFS) measurements of the heterowire at the Cu K-edges exhibit characteristics the same as those of Cu foil compared to Cu 2 O and CuO in the XANES spectra (Figure 1i), indicating that the Cu NWs remain metallic in Cu NWs@ZIF-8.The Fourier transform Cu K-edge EXAFS curve of Cu NWs@ZIF-8 (Figure 1j) displays intense peaks located around 2.2 Å corresponding to the Cu-Cu bond, confirming the effective protection of the Cu NWs from oxidation by the MOF cover, which guarantee the perfect long-term stability of these dual-channel catalysts.
The morphology of Cu NWs@ZIF-8 was characterized through a combination of microscopy techniques (SEM and TEM).The SEM images show that the covered Cu NWs are well dispersed and that a layer of ZIF-8 has been uniformly grown on the Cu NWs to form a skin-core structure (Figure 1d,e and Figure S6, Supporting Information).A secondary electron imaging (SEI) image demonstrates the smooth surface of a Cu NWs@ZIF-8 composite composed of polycrystalline ZIF-8 (Figure 1f and Figure S7, Supporting Information).In the TEM images, the Cu NWs core and the ZIF-8 shell are clearly illustrated (Figure 1e and Figure S7, Supporting Information).Elemental mapping analysis was carried out by means of TEM in scanning transmission electron microscopy (STEM) mode to reveal the spatial distribution of different elements in Cu NWs@ZIF-8 (Figure 1f).Extending outward from the center along the radial direction of the nanowire, the Cu elemental mapping represents Cu NWs, and Zn element mapping (including C and N elemental mapping) represents the ZIF-8 layer.The thickness of the ZIF-8 shell is ≈40 nm.

Electrocatalytic Reduction of CO 2 to CO with the Heterowire Cu NWs@ZIF-8
The electrochemical characterization of the dual-channel heterowire catalyst Cu NWs@ZIF-8 and its electrocatalytic CO 2 reduction performance was carried out in a commercial gas-tight H-type flow cell with a standard three-electrode system at room temperature under atmospheric pressure.A glassy carbon electrode was used as the working electrode, an Ag/AgCl electrode with a saturated KCl filling solution was used as the reference electrode, and a platinum wire electrode was used as the counter electrode.Linear sweep voltammetry (LSV) in a CO 2 -saturated 0.1 m potassium bicarbonate (KHCO 3 ) electrolyte (pH = 6.8) was used to initially determine the electrochemical activity and the onset reduction potential of the catalyst.Current densities were normalized by the geometric surface.Fifty cycles of cyclic voltammetry (CV) sweeps were conducted before LSV was performed.The LSV curves of Cu NWs and Cu NWs@ZIF-8 (Figure 2c) indicate that the onset potential was more negative than −0.60 V versus reversible hydrogen electrode (RHE).Considering that a considerable current density was required to evaluate the catalytic TEM image of as-synthesized Cu NWs@ZIF-8.f) HAADF-STEM image and elemental maps (C, N, Cu, Zn) of Cu NWs@ZIF-8.g) Synchrotron PXRD ( = 0.77471 Å) of as-prepared Cu NWs@ZIF-8.h) CO 2 uptake amount of ZIF-8 and Cu NWs@ZIF-8 at 298 K. i) Normalized Cu K-edge XANES spectra of Cu NWs@ZIF-8.j) R-space Cu K-edge EXAFS spectra of Cu NWs@ZIF-8.activity of the catalyst, a reduction potential range of −0.7 to −1.3 V RHE was used in the electrocatalytic reduction of CO 2 .As shown in Figure 2a, Cu NWs exhibited a wide distribution of reduction products under the applied reduction potential.When the reduction potential was more positive than −0.9 V RHE , H 2 was generated with a faradaic efficiency (FE) over 40%, and a mixture of CO, C 2 H 4 , and CH 4 in the gas phase and HCOOH in the liquid phase were obtained from the CO 2 RR, with a poor total FE.Although the HER was suppressed under a more negative reduction potential (−1.0 V RHE ), the minimum FE still reached 20% at −1.1 V RHE , and the total FE of the CO 2 RR was ≈95%, with a product distribution of 22.3% C 2 H 4 , 15.0% CH 4 , 7.7% CO and a small amount of CH 3 OH.This product distribution is similar to that obtained for the CO 2 RR on polycrystalline metallic nano-copper reported previously. [55,56]The LSV curve of ZIF-8 (Figure S8, Supporting Information) shows negligible activity to electrocatalytic CO 2 reduction.
The same reduction potential range was used to evaluate the activity of the Cu NWs@ZIF-8 according to the comparable onset reduction potentials of Cu NWs (Figure 2b).As clearly shown in Figure 2b, both the FE and selectivity of CO 2 RR were greatly improved for the composite Cu NWs@ZIF-8 compared with Cu NWs.The onset reduction potential to CO was significantly reduced, and an FE of CO more than 60% was obtained at −0.7 V RHE .CO was the only carbon product in the entire test potential range.The FE of CO was increased to 65.2-91.3%,and the competitive HER was effectively suppressed as low as 5.2%.At −1.1 V RHE , ≈100% total FE and 91.3% CO were achieved (Figure 2b).It is worth noting that when the current density was normalized to the unit mass of Cu (Figure 2d; Table S1, Supporting Information), the equivalent CO yield was increased by 65 times along with the perfect selectivity.Subsequently, another important parameter for performance of the electrocatalytic reduction of CO 2 catalyst, the long-term stability of the heterowire catalyst was evaluated at −1.1 V RHE .The HER was well suppressed during the whole measurement, and Cu NWs@ZIF-8 exhibited excellent cycle stability for both FE and selectivity of CO as well as the total current density even after 24 hours (Figure S9, Supporting Information), which was superior than all the other Cu-based CO 2 RR electrocatalysts (2-15 h; Table S2, Supporting Information), indicating the reliable protective effect of the ZIF-8 cover around the Cu NWs.
SEM images show that the catalyst maintains its original morphological structure after the reaction (Figure S10, Supporting Information).In contrast, Cu NWs is easily inactivated under the same reaction conditions, with the extension of the reaction time, the FE of H 2 increased rapidly after 7 h (Figure S11, Supporting Information).This may be caused by surface reconstruction of the Cu NWs during the reaction. [55,57]The SEM images (Figure S12, Supporting Information) of the Cu NWs after the reaction shows that there are many small particles on the surface compared to the smooth surface of the fresh sample, and the carbon products of CO change much at a low level.This proves that our synthesis strategy is very effective in improving activity and maintaining stability in CO 2 RR.
Density functional theory (DFT) calculations were performed to insight into the activity and selectivity of the Cu NWs@MOF dual-channel catalytic system for CO 2 RR.During the electrocatalytic reduction of CO 2 by Cu NWs@ZIF-8, we speculate that the catalytic active center of the 1D composite is located at the interface between Cu NWs and ZIF-8 since the activity of ZIF-8 is negligible compared to that of Cu NWs (Figure S8, Supporting Information).Therefore, the binding energy of key reaction intermediates *COOH and *CO on Cu(100) of Cu NWs and on the interface of Cu NWs and ZIF-8 were calculated, respectively.Figure 3a,b shows the optimized structures used in DFT calculations and the energetically most favorable adsorption configurations of the intermediates *COOH and *CO on Cu NWs@ZIF-8.As shown in Figure 3c, the formation of key intermediates *COOH and *CO in Cu NWs@ZIF-8 composite system is the dominant path compared with the Cu(100) surface (Figure S13a-g, Supporting Information), and more importantly, the desorption of *CO to form CO is thermodynamically favorable.Consistent with the experimental measurements, the theoretical calculation results confirm the electron and molecule transmission between Cu NWs and the MOF cover, and clearly illustrate the high activity and excellent CO selectivity of the 1D nanosized Cu NWs@MOF dualchannel catalyst originating from the cooperation of the electron channel and the molecule channel.Benefiting from the fine conductivity of the Cu NWs electron channel directionally transmitted and continuously supplied electrons to the CO 2 RR process, while the MOF cover acted as the multifunctional molecule channel to protect the Cu NWs, auto-capture CO 2 molecules at the interface between Cu NWs and MOF, and maintain a proper CO 2 concentration surrounding the Cu NWs based on the good stability and the high BET surface area of ZIF-8 (Figure S14 and Table S1, Supporting Information), the heterowire Cu NWs@ZIF-8 catalyst performed as the nanosized dual-channel composite catalyst to produce relative pure CO with all-round advantages including high efficiency, good conversion rate, promising selectivity and tremendous long-term stability.

Dual-Channel Heterowire Catalysts to Control the CO/H 2 Ratio of Syngas Products by Co 2+ Doping
Syngas, one of the most important resources as both an energystorage fuel and a building block for the synthesis of various industrial products, is a gaseous mixture of CO and H 2 , [58,59] Since the electrochemical reduction of CO 2 is generally performed in aqueous media, HER from the reduction of water or protons is in inevitable rivalry with the CO 2 conversion.Hence, combining CO 2 reduction and HER for the production of syngas is believed as the intrinsic nature and one of the most plausible approach for electrocatalytic reduction of CO 2 . [5,60]In view of the unique advantages of the ZIF-8 channel in Cu NWs@ZIF-8 for reduction of CO 2 to CO, we attempted to switch the CO/H 2 ratio to produce syngas from the CO 2 RR by adjusting the composition of the molecule channel.Following the synthesis of Cu NWs@ZIF-8, Co 2+ ions together with Zn 2+ were introduced during the in-situ coating of the ZIF cover Considering both energy consumption and current density, the activity of these dual-channel catalysts were evaluated for the CO 2 RR at a reduction potential of −1.1 V RHE .The product distribution, FE values and current densities are summarized in Figure 5 and Figure S36 (Supporting Information).As the ratio of Co 2+ in the MOF channel increased, the main component of the CO 2 RR products changed from CO for Cu NWs@ZIF-8 to syngas.When the Zn/Co ratio of the catalyst was 0.5:0.5 (Cu NWs@Zn 0.5 Co 0.5 -ZIF), the CO/H 2 ratio of the product was ≈1:1 with a promising FE of 97.5%.Further changing the Zn/Co ratio of the hybrid catalysts (Cu NWs@Zn 0.4 Co 0.6 -ZIF, Cu NWs@Zn 0.3 Co 0.7 -ZIF, Cu NWs@Zn 0.2 Co 0.8 -ZIF), the CO/H 2 ratio in syngas was adjusted from 1:1 to 1:3 with no other by-product and maintained FE.When the metal ion in the ZIF cover was totally Co 2+ , the heterowire with a ZIF-67 cover, the Co-containing analogue of ZIF-8, resulted a CO/H 2 ratio of 1:3.2, confirming that Co 2+ was active to HER and thus acted as an assistant for generating an appropriate amount of H 2 (Figure S37-S38, Supporting Information) in Cu NWs@Zn x Co 1−x -ZIF. [61]As a result, a syngas series with different compositions could be selectively produced through CO 2 RR from these switchable heterowire catalysts by either adjusting the MOF components of the molecule channel or controlling the electric parameters of the electron channel, successively satisfying different synthetic requirements for a vast range of essential industrial products. [58,59,62]

Dual-Channel Mott-Schottky Heterowire for Photocatalytic Reduction of CO 2
In order to further extend the newly developed dual-channel heterowire catalyst family, we raised Cu 3 (BTC) 2 (BTC = 1,3,5benzene tricarboxylate) (HKUST-1), a semiconducting MOF, [63] as the cover of Cu NWs instead of ZIF for introducing the photocatalytic activity besides the molecule and electron channels.XRD and SEM (Figure 6a,b) data clearly confirm the successful setup of Cu NWs@HKUST-1.The activity of Cu NWs@HKUST-1 for photocatalytic reduction of CO 2 (1 bar) was carried out at room temperature under visible-light irradiation ( > 420 nm) (Figure S39, Supporting Information) with water vapor as the only electron donor.Remarkably, promising photocatalytic activity of Cu NWs@HKUST-1 on CO 2 reduction was obtained with the yield of CO up to 176.2 μmol h −1 g −1 in 5 hours accompanied by a small amount of C 2 H 4 , and more importantly, the selectivity of CO reached 92%.Except for CO and C 2 H 4 , no other carbon product was detected in both gas and liquid phases.As a comparison, when solely Cu NWs or HKUST-1 was used as catalyst, only a small amount of C 2 H 4 (15.9 μmol h −1 g −1 ) or CO (32 μmol h −1 g −1 ) was observed, respectively.In addition, as control experiments, no product was detected when using Ar instead of CO 2 , confirming CO 2 as the carbon source of CO and C 2 H 4 .If Cu NWs and HKUST-1 were simply physically mixed as catalysts, a much higher H 2 production was observed, indicating that the catalytic active site is located at the interface between Cu NWs and HKUST-1.
To further investigate the molecule and electron exchange between the two channels and the catalytic mechanism of Cu NWs@HKUST-1, Mott-Schottky measurements were conducted at frequencies of 1000, 1500, and 2000 Hz (Figure S40, Supporting Information).The positive slope of the Mott-Schottky patterns illustrated HKUST-1 as a typical n-type semiconductor.The flat band position (V fb ) determined from the intersection of Mott−Schottky plots was ≈−0.84 V vs Ag/AgCl.Therefore, the Fermi level of HKUST-1 (3.86 eV vs AVS) was given by EF = V fb (vs SHE) e-4.44 eV. [64]Since the bottom of LUMO in n-type semiconductors should be approximately equal to the flatband potential, the LUMO of HKUST-1 was estimated as 3.66 eV vs AVS.The band gap of HKUST-1 was thus estimated as 2.41 eV from the Tauc plot (Figure S41, Supporting Information), and its HOMO was accordingly calculated as 6.07 eV vs AVS.Based on the energy levels of HKUST-1 in the composite catalyst, the schematic depicting for the band alignments and charge flow at the metal-MOF interfaces involved in the system has been illustrated in Figure 6d.First, HKUST-1 undergoes electron-hole separation under visible light irradiation.Since the Fermi energy level of HKUST-1 (3.86 eV) is lower than the work function of Cu NWs (4.65 eV), [65] the electrons in the conduction band of HKUST-1 are delivered to Cu NWs driven by Cu NWs-MOF Schottky barrier, and the interfacial electron exchange generates an electron-rich Cu NWs together with an electron depletion region on the HKUST-1 side.Finally, the electron-rich Cu NWs results in the highest photoactivity of Cu NWs@HKUST-1 for reducing CO 2 to CO.It is worthy to mention that as far as we know this is the first example of an 1D/1D type Mott-Schottky heterojunction. [65]

Discussion
For the electrocatalytic CO 2 reduction in a typical H-type flow cell, the catalytic process mainly includes the following four steps: 1) the continuously bubbling gas phase CO 2 dissolves in the electrolyte; 2) the dissolved CO 2 in the electrolyte diffuses to the surface of the catalyst on the electrode; 3) CO 2 molecules are adsorbed and activated on the catalyst surface to complete the catalytic reaction; and 4) the reaction products are desorbed and dissociated. [9]When the MOF cover of the Cu NWs@MOFs heterowire is insulated such as ZIF-8, the electron channel Cu NW directionally transmits the electrons and may form "micro current" along the Cu NW due to the fine conductivity of copper, as confirmed by the electrochemical impedance spectroscopy (EIS) test of Cu NWs@ZIF-8 in comparison with ZIF-8 (Figure S42, Supporting Information).Electrons flowing through the Cu NWs are continuously supplied to the adsorbed CO 2 molecules, which may have greater opportunity to be reduced in a "long journey" of ≈10 μm along the Cu NWs, leading to much higher electronic utilization efficiency compared with Cu nanoparticles.From a point of view, the dual-channel Cu NWs@ZIF readily mimic the electric cable consisting of conduction pith and insulation cover, that well regulates the electronic transmission and suppresses the microcurrent power loss.
As the molecule channel, the MOF cover controls the entrance of the reactant molecules and the departure of the product molecules, and brings efficiency, selectivity and stability for the dual-channel electrocatalytic system: i) the MOF cover significantly improved the dispersion of the Cu NWs and hindered the ultralong Cu NWs from intertwining.ii) The coating of MOF cover significantly enhanced the BET surface area from only 23 m 2 g −1 of the Cu NWs to as high as 800 m 2 g −1 of Cu NWs@ZIF-8 (Figure S13 and Table S1, Supporting Information), providing a much higher electrochemical active surface area (ECSA) (Figure S43, Supporting Information), which is conducive to obtain a superior CO 2 RR FE and a tenfold increase in CO 2 converting rate.iii) As illustrated by the hydrophilia tests (Figure S44, Supporting Information), Cu NWs, Cu NWs@ZIF-8 and Cu NWs@ZIF-67 shows contact angle of 129.6°, 28.7°, and 26.5°w ith water, respectively, indicating the hydrophilic character of the MOF cover, which is benefit for transporting electrolytes containing CO 2 (aq) and protons toward the interface between the Cu NWs and the MOF for the continuous reduction of CO 2 .iv) Furthermore, when using the semiconducting MOF cover such as HKUST-1 for CO 2 , besides molecule channel MOF even acts as a "fiber converter" to absorb photons and supply electrons to Cu NWs through the Schottky-effect-induced electron flow at the interface between Cu NWs and HKUST-1, and the electron enrichment of Cu NWs can be further enhanced via photoinitiated charge separation at the Cu NWs-HKUST-1 interface to significantly promote the efficient photoreduction of CO 2 to CO.

Conclusion
In conclusion, a series of 1D dual-channel Cu NWs@MOFs catalysts were constructed by rationally coating MOFs on the sur-face of Cu NWs, in which Cu NWs acted an electron channel facilitated electronic transmission while MOFs acted as molecule and/or photon channels to provide an appropriate transport path for both reactants/intermediates and photoelectric conversion.Excellent Faraday efficiency, drastically suppressed HER, and the highest stability of the Cu-based CO 2 RR electrocatalysts, were obtained by the heterowire catalyst for the electrocatalytic conversion of CO 2 .Moreover, the distribution of the reduction product could be precisely regulated from CO to syngas with different ratio of CO/H 2 by adjusting the composition of the MOF, which gave the first heterometallic MOF covered 1D composite, Cu NWs@Zn x Co 1−x -ZIF.When further switching the MOF cover from insulator to a semiconductor, the first 1D/1D type Mott-Schottky heterojunction, Cu NWs@HKUST-1, was constructed.Attributed to the Schottky-effect induced electron flow at the 1D/1D metal-semiconductor interface, Cu NWs@HKUST-1 exhibits superior catalytic activity in the photocatalytic reduction CO 2 under visible light irradiation without additional sacrificial agent other than water.The above findings provide a reliable and universal solution for the design of CO 2 electro-/photoreduction catalysts.Considering the diversity of MOFs, it is foreseeable that our strategy will play an increasingly important role in both CO 2 management and energy conversion in the future.O) was purchased from Acros Organics.KOH and hexane (99.9%) were purchased from Fisher Chemical.All the chemicals were used without purification.The deionized water (18.2MΩ cm) was used to make the solutions.

Experimental Section
Preparation of Cu NWs: Cu NWs were prepared by an improved hydrothermal method according to the literature.In a typical procedure, 40 mL of an aqueous solution containing 85 mg of CuCl 2 •2H 2 O, 500 mg of HDA, and 111 mg of d-(+)-glucose was vigorously stirred overnight under 45 °C to obtain an emulsion mixture with light-blue color, followed by sonication for 20 mins.The solution was transferred into a Teflon-lined stainless-steel autoclave (100 mL).After 12 h of hydrothermal reaction at 120 °C, the solution was cooled to room temperature and a brown mixture was obtained.The solid product obtained by centrifugation (10 000 rpm, 5 min) and washed three times with water, ethanol, and hexane respectively.Finally, the Cu NWs were stored in 8 mL absolute ethanol.
Preparation of ZIF-8: ZIF-8 was prepared according to the literature. [66,67]Briefly, dissolve 372 mg Zn(NO 3 ) 2 •6H 2 O in 50 mL methanol to obtain solution A, and dissolve 164 mg 2-methylimidazole in 50 mL methanol to obtain solution B. Then solution A and solution B were mixed and allowed to react at room temperature for 24 h without stirring.The solid product was collected by centrifugation, washed several times with methanol, and vacuum-dried overnight at 70 °C.
Preparation of ZIF-67: ZIF-67 was prepared according to literature reported method.In brief, dissolve 364 mg Co(NO 3 ) 2 •6H 2 O in 50 mL methanol to obtain solution C, and dissolve 164 mg 2-methylimidazole in 50 mL methanol to obtain solution D. Then solution C and D were mixed and allowed to react at room temperature for 24 h without stirring.The solid product was collected by centrifugation, washed several times with methanol, and vacuum-dried overnight at 70 °C.isotherms at 77 K (ASAP 2020 PLUS Analyzer, Micromeritic).CO 2 uptake amount were determined at 298 K (ASAP 2020 V4.03, Micromeritic).Metal loadings in the catalysts were determined by inductively coupled plasma optical emission spectrometry (ICP-OES, SpectroBlue, Spectro).X-ray photoelectron spectroscopy (XPS, Thermo ESCALAB 250XI, ThermoFisher), The charging effect was corrected using the C1s level (284.8 eV) as the reference.XAFS spectra at the copper K edge and were collected at Australian Synchrotron, ANSTO.Synchrotron powder X-ray diffraction (Synchrotron PXRD) measurements were measured at PD beamline, Australian Synchrotron, ANSTO with a wavelength of  = 0.77471 Å. UV-vis absorption diffuse reflectance spectra (UV-vis DRS) were recorded on a Cary 100 UV-vis spectrophotometer using an integrating sphere accessory.
Preparation of Working Electrode: 4 mg of freshly prepared dried Cu NWs were dispersed in 1 mL absolute ethanol and ultrasonication for 1 h.Subsequently, 30 μL of Nafion (5 wt%) was added and ultrasonication for another 10 min.20 μL catalyst ink was dropped onto the surface of a precleaned glassy-carbon electrode (diameter, 3 mm; area, 0.07 cm 2 ) using pipettor and dried under ambient air.For catalysts ZIF-8, ZIF-67, Cu NWs@ZIF-8, and Cu NWs@ZIF-67, the same method was used to prepare electrodes.The Cu content of catalysts Cu NWs@ZIF-8 and Cu NWs@ZIF-67 is 22.2 and 20.7 wt% measured by ICP-OES, respectively.
Electrochemical Measurements: All electrochemical measurements were carried out on an Autolab PGSTAT302N potentiostat (Metrohm Autolab) at room temperature.Electrochemical measurements and electrochemical reduction of CO 2 experiments were conducted by using a customized gastight two-chamber H-type cell with a standard three-electrode system.Glassy carbon electrode coated with catalyst was used as working electrode, Ag/AgCl electrode with a saturated KCl filling solution regularly calibrated by a reversible hydrogen electrode (RHE) was used as the reference electrode and a Pt coil from was used as a counter electrode.Working and reference electrodes were fixed in one chamber and the counter electrode was fixed in the other chamber.Two chambers (Cathode and anode) were separated by a Nafion ion exchange membrane (Nafion 117 membrane, Alfa) and the electrolyte in both chambers are 50 mL 0.1 m KHCO 3 .For cyclic voltammetry (CV) and linear sweep voltammetry (LSV) test, the electrolyte was bubbled with CO 2 or N 2 for at least 60 min to get the electrolyte saturated, electrochemical impedance spectroscopy (EIS) was conducted at 0 V vs open-circuit potential from 0.01 Hz to 100 kHz with an amplitude of 5 mV.The LSV was performed at a scan rate of 10 mV s −1 to qualitatively evaluate the catalytic activity and define the constant voltage applied to study the product distribution at each potential.
The Electrocatalytic CO 2 Reduction: CO 2 RR performance of catalysts were evaluated in a H-type flow cell.Before CO 2 RR, the electrolyte was bubbled CO 2 (Airgas, 99.99%) for 60 min to reach saturation.During the measurement, CO 2 was continuously purged through a mass flow controller (HORIBA, Japan) into cathode compartment with a flow rate of 10 sccm with stirring at 1000 rpm.The FE was measured by using chronoamperometry for 2-3 h at each applied potential.Gas-phase products of CO 2 reduction were quantified by a gas chromatography (GC, Agilent 8860) equipped with equipped with thermal conductivity detectors (TCD) and flame ionization detector (FID).A series of standard gas mixtures (H 2 , CO, CH 4 , C 2 H 4 , and C 3 H 6 , Airgas) were used to establish the calibration curves.1H-NMR (Bruker, 400 MHz) was used to detect the liquid phase product.650 μL of the electrolyte after electrolysis was mixed with 300 μL of D 2 O and 50 μL dimethyl sulfoxide (DMSO) was added as an internal standard.All quantitative data is the average value of three measurements under the same test conditions.
Photoelectrochemical Measurements: The Mott-Schottky plots (MS) was measured with Autolab PGSTAT302N potentiostat (Metrohm Autolab) by a traditional three-electrode system.Specifically, a platinum plate was used as the counter electrode and an Ag/AgCl electrode was used as the reference electrode.The preparation of the working electrode was as follows: 5 mg prepared sample was dispersed in the solution of Nafion/ethanol (10 μL/1 mL) and ultrasonicated for 1 h.Then, 80 μL of the mixture was dropped onto an indium tin oxide (ITO) glass substrate with an area of 1.0 cm 2 .The obtained electrode was dried at 120 °C under a vacuum overnight.The electrolyte was 0.5 m sodium sulfate aqueous solution.A 300 W Xe lamp with a 420 nm cutoff filter was used as the light source.
The Photocatalytic CO 2 Reduction: The photocatalytic CO 2 reduction was performed in a 125 mL homemade photoreactor, which was irradiated with a 300 W Xe lamp (HSXF300, Beijing NBET Technology Co., Ltd.) equipped with a 420 nm cutoff filter at 25 °C.Typically, 5 mg catalyst samples were spread on a quartz filter membrane which was placed on the sample holder built in the center of the reacting vessel, then 5 mL pure water was added below the holder, avoiding direct contact with the samples.Then the reactor was filled with 0.1 MPa of high-purity CO 2 .The gas products after the reaction were quantitatively analyzed by Agilent 8860 gas chromatography equipped with thermal conductivity detectors TCD and FID detector.Helium was used as the carrier gas to detect H 2 , and nitrogen was the carrier gas used to detect other gas products.The liquid phase product was detected by 1H-NMR (Bruker, 400 MHz).
[71][72] The generalized gradient approximation was used with the Perdew-Burke-Ernzerhof [73] exchange-correlation functional.The projector-augmented wave method [74,75] was utilized to describe the electron-ion interactions and the cut-off energy for the plane-wave basis set was 450 eV.To illustrate the long-range dispersion interactions between the adsorbates and catalysts.Brillouin zone integration was accomplished using a 3×3×1 Monkhorst-Pack k-point mesh.All the adsorption geometries were optimized using a force-based conjugate gradient algorithm.For the modelling of Cu(100), the crystal structure was optimized; Cu(100) were modelled with a periodic three-layer p(3×3) model with the two lower layers fixed and the two upper layers relaxed.Cu NWs@ZIF-8 was modelled with a periodic three-layer p(10×10) model.In order to reduce the amount of calculation, a window of the ZIF-8 cage was chosen to grow on the Cu(100) surface for modeling.The atomic coordinates of the optimized models are provided in Figure S12 (Supporting Information).

Figure 2 .
Figure 2. Catalytic activity and stability of the dual-channel heterowire Cu NWs@ZIF-8 for electrocatalytic reduction of CO 2 .a) FE of the main products for Cu NWs under different applied reduction potentials.b) FE of the main products for Cu NWs@ZIF-8 under different applied reduction potentials.c) LSV of Cu NWs@ZIF-8 in N 2 and CO 2 saturated 0.1 m KHCO 3 electrolyte.d) Partial current densities toward CO and H 2 for Cu NWs@ZIF-8 in 0.1 m KHCO 3 .

Figure 3 .
Figure 3. DFT calculations.a) DFT optimized structure of intermediate *COOH adsorbed on Cu NWs@ZIF-8.b) DFT optimized structure of intermediate *CO adsorbed on Cu NWs@ZIF-8.The orange, pink, grey, blue and red spheres represent Cu, Zn, C, N, and O atoms, respectively.c) Calculated binding energies (in eV) of key reaction intermediates on Cu(100) and Cu NWs@ZIF-8 at a potential of 0 V. d) Calculated binding energies (in eV) of key reaction intermediates on Cu(100) and Cu NWs@ZIF-8 at a potential of −1.1 V RHE .