UPGRADING OF CO TO C3 PRODUCTS USING MULTI-METALLIC ELECTROREDUCTION CATALYSTS WITH ASSYMETRIC ACTIVE SITES

Abstract

The present disclosure relates to electrocatalysts for electroreduction of a carbon-containing gas to produce n-propanol, for example. The electrocatalyst includes a multi-metallic material comprising a primary metal, such as Cu, and a metal dopant, such as Ag, selected and distributed to provide asymmetric active sites that include neighbouring atoms of the primary metal having distinct electronic structures to promote C2-C1 coupling. The electrocatalysts can be bimetallic or bimetallic, for example. The disclosure also relates to manufacturing and using the electrocatalysts, which can be used as a cathodic catalyst to convert CO or CO2 into multi-carbon products.

Description

TECHNICAL FIELD

The technical field generally relates to catalytic methods for carbon dioxide (CO₂) and monoxide (CO) reduction, and more particularly to electrocatalysts composed of metallic material such as Cu doped with Ag or with Ag and Ru, and associated methods of manufacture and use in electrochemical reduction for the production of C3 products.

BACKGROUND

The electrosynthesis of C3 products from carbon dioxide (CO₂) and carbon monoxide (CO) addresses the need for long-term storage of renewable electricity. Unfortunately, present-day performance remains below that needed for practical applications. There is a need for improved techniques and catalyst materials for efficient electrochemical reduction of gases, such as CO and CO₂, and related methods and systems of producing chemical compounds, such as C3 compounds.

SUMMARY

Various implementations, features and aspects of the technology are described herein and relate to electroreduction electrocatalysts for converting a carbon-containing gas to produce products, such as n-propanol, where the electrocatalysts include a multi-metallic material comprising a primary metal, such as Cu, and a metal dopant, such as Ag and optionally a second metal dopant, selected and distributed to provide structures that promote C2-C1 coupling. The technology also relates to manufacturing of the electrocatalysts and their use for the electroreduction of CO or CO₂, for example.

In some implementations, there is provided an electrocatalyst for electroreduction of a carbon-containing gas to produce C3 products, the electrocatalyst comprising a multi-metallic material comprising a primary metal and a metal dopant selected and distributed to provide asymmetric active sites that include neighbouring atoms of the primary metal having distinct electronic structures to promote C2-C1 coupling.

Preferably, the primary metal is Cu, the metal dopant is Ag, and the multi-metallic material is a bimetallic material. The metal dopant can include Ag and a second dopant metal and the multi-metallic material is a trimetallic material.

In some implementations, the carbon-containing gas comprises or is CO, CO₂, or a mixture thereof. In some implementations, the C3 product is n-propanol.

In some implementations, multi-metallic material comprises the primary metal doped with the metal dopant using galvanic replacement. The metal dopant can be present in the primary metal in a doping concentration of 2 wt % to 9 wt %, in a doping concentration of 3 wt % to 8 wt %, in a doping concentration of 3wt % to 5 wt %, in a doping concentration of 3.5 wt % to 4.5 wt %, or in a doping concentration of approximately 4 wt %, measured with XPS.

In some implementations, the electrocatalyst is provided in the form of bimetallic or trimetallic nanoparticles. The bimetallic or trimetallic nanoparticles can have an average size between about 20 nm and about 200 nm, or between about 50 nm and about 200 nm, between about 70 nm and about 150 nm, or between about 90 nm and 130 nm, measured based on SEM or TM imaging. The bimetallic or trimetallic nanoparticles can be generally spheroid in shape, determined from SEM or TM imaging.

In some implementations, the electrocatalyst is formed as a deposited catalyst layer on a first side of gas diffusion membrane, wherein the deposited catalyst layer is configured to be in direct contact with an electrolyte and wherein a second opposed side of the gas diffusion membrane is configured to be in direct contact with the carbon-containing gas.

In some implementations, there is provided an electrocatalyst for electroreduction of a carbon-containing gas to produce a C3 product, the electrocatalyst comprising a bimetallic or trimetallic material comprising a copper (Cu) and at least one metal dopant in a doping concentration of 2 wt % to 9 wt %, or 3 wt % to 8 wt %, or 3wt % to 5 wt %, or 3.5 wt % to 4.5 wt % or approximately 4 wt %, measured with XPS. The metal dopant can include or be Ag. The metal dopant can include a primary dopant and a secondary dopant. In some implementations, the primary dopant is Ag and/or the secondary dopant is Ru.

In some implementations, there is provided a method of fabricating an electrocatalyst composed of a primary metal and at least one metal dopant for electroreduction of a carbon containing gas into C3 products, such as n-propanol, comprising galvanic replacement.

The method can include providing a layer of Cu particles on a substrate to provide a coated substrate; immersing the coated substrate in an Ag containing aqueous solution to induce doping and form an Ag-doped multimetallic catalyst material supported by the substrate; and removing the coated substrate from the solution, the coated substrate comprising a layer of the Ag-doped multimetallic catalyst material. In some implementations, the Cu particles comprise Cu nanoparticles. In some implementations, the providing of the layer of Cu particles on the substrate is performed by spray coating to form the coated substrate. In some implementations, the Ag containing aqueous solution is an AgNO₃ aqueous solution. In some implementations, a second dopant metal is incorporated to form the multimetallic catalyst material supported by the substrate.

In some implementations, there is provided a process for electrochemical production of a C3 multi-carbon compound from a carbon-containing gas, comprising: contacting the carbon-containing gas and an electrolyte with an electrode comprising the electrocatalyst as defined herein or as manufactured by the method as defined herein, such that the carbon-containing gas contacts the electrocatalyst; applying a voltage to provide a current density to cause the carbon-containing gas contacting the electrocatalyst to be electrochemically converted into the C3 multi-carbon compound; and recovering the C3 multi-carbon compound.

In some implementations, the C3 multi-carbon compound is an alcohol, such as propanol which may be n-propanol. In some implementations, the electrolyte comprises an alkaline compound. The electrolyte can include KOH and/or other alkaline solutions. In some implementations, carbon-containing gas comprises or is CO, CO₂ or both.

In some implementations, there is provided a system for CO and/or CO₂ electroreduction to produce a multi-carbon compound, comprising: an electrolytic cell configured to receive a liquid electrolyte and CO and/or CO₂ gas; an anode; a cathode comprising an electrocatalyst as defined herein or as manufactured by the method as defined herein; and a voltage source to provide a current density to cause the CO and/or CO₂ gas contacting the electrocatalyst to be electrochemically converted into the multi-carbon compound.

DESCRIPTION OF DRAWINGS

The Figures describe various aspects and information regarding the technology.

FIG. 1 | DFT calculations on C1-C1 and C1-C2 coupling. DFT calculated reaction barriers (Ea) for C1-C1 and C1-C2 coupling on screened M-doped Cu systems (M=Ag, Au, Ru, Rh, and Pd). The geometries of M-doped Cu surface, C1, C2, and C3 on M-doped Cu are shown with the corresponding labels, respectively. Cu, M, C, and O are illustrated as orange, light blue, grey, and red balls, respectively, while water molecules are shown as lines.

FIG. 2 | Structural and compositional analyses of Ag-doped Cu catalyst. a, High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image taken from a single particle. Scale bar, 20 nm. b, Atomic-resolution HAADF-STEM image taken from the edge of a nanoparticle marked by a box in (a). Inset, the corresponding Fourier transfer image. Scale bar, 2 nm. c, HAADF-STEM image of an Ag-doped Cu nanoparticle and the corresponding EELS elemental mappings of Cu, Ag, and O. Scale bar, 20 nm. d,e, WAXS map (d) and the corresponding sector-average of WAXS map without smooth (e) for Ag-doped Cu GDE. f,g, High-resolution Cu 2p (f) and Ag 3d (g) spectra of Ag-doped Cu GDE.

FIG. 3 | CO electroreduction performance and operando structural characterizations of Ag-doped Cu catalyst in flow cell. a, FEs of n-propanol (n-PrOH), ethanol, acetate, and ethylene on Ag-doped Cu and pristine Cu catalysts under different potentials. b, Comparison of n-propanol FEs on Ag-doped Cu and Cu GDE under different potentials, as well as partial current density of n-propanol formation on Ag-doped Cu GDE. Error bars represent the standard deviation based on three separate measurements. c, Operando Cu K-edge XANES spectra of Cu GDE following 10 s and Ag-doped Cu GDE following 10, 20, 30, 80, 600, 1200, 1800, and 2400 s at −0.46 VRHE during CORR. Bulk Cu foil, CuO, and Cu2O are listed as references. d, Operando Cu K-edge EXAFS spectrum of Ag-doped Cu GDE following 10 s at −0.46 VRHE during CORR. Bulk Cu foil was listed as a reference. Inset, an enlarged view of the section from 0.7 to 5.3 Å for Ag-doped Cu GDE.

FIG. 4 | CO electroreduction performance on different types of Ag-doped Cu catalysts and operando structural characterizations in flow cell. a, FEs of n-propanol, ethanol, acetate, and ethylene on Cu—Ag-20 min, Cu—Ag-2 h, and Cu GDE under different potentials. b, Comparison of n-propanol FEs on Cu—Ag-20 min, Cu—Ag-2 h, and commercial Cu GDE. Error bars represent the standard deviation based on three separate measurements. c, Operando Cu K-edge XANES spectra of Cu—Ag-20 min, Cu—Ag-2 h, and Cu GDE following 10 s at −0.46 VRHE. Bulk Cu foil, Cu2O, and CuO are listed as references.

FIG. 5 | Geometries of CO dimerization on Ag-doped Cu surface. a-c, Side views of initial state (a), transition state (b), and final state (c). d-f, Top views of initial state (d), transition state (e), and final state (f). Light blue balls stand for silver atoms. The distance of CO molecules at the transition state of CO dimerization on Ag-doped Cu is 1.908 Å. This notation is used throughout in the Supplementary Information section.

FIG. 6 | Geometries of C1 and C2 coupling on Ag-doped Cu surface. a-c, Side views of initial state (a), transition state (b), and final state (c). d-f, Top views of initial state (d), transition state (e), and final state (f). Light blue balls stand for silver atoms.

FIG. 7 | DFT calculations on C1-C1 and C1-C2 coupling. a, Activation energies of C1-C1 and C1-C2 coupling on Cu, Cu with strain, and Ag-doped Cu. Cu with strain is the Cu surface with the same bond length as Ag-doped Cu but without Ag substitution. b, Two types of neighbouring Cu atoms labeled as a and b on Ag-doped Cu surface.

FIG. 8 | SEM images of catalysts. a, Pristine Cu catalyst. b, Ag-doped Cu catalyst.

FIG. 9 | Structural analysis of an Ag-doped Cu nanoparticle. a, HAADF-STEM image of the same Ag-doped Cu nanoparticle in FIG. 2c. b,c, Atomic-resolution HAADF-STEM images taken from the edges of a nanoparticle marked by a box in (a). The analysis of atomic-resolution HAADF-STEM images demonstrated the both (111) and (100) facets were exposed on Ag-doped Cu nanoparticle.

FIG. 10 | Compositional analysis of an Ag-doped Cu nanoparticle. a, HAADF-STEM image of the same Ag-doped Cu nanoparticle in FIG. 2c. b, Overlap of the corresponding EELS elemental mappings of Cu, Ag, and O in FIG. 2c.

FIG. 11 | Structural characterizations of GDE. a, WAXS map of Cu GDE. b, XRD patterns for Cu and Ag-doped Cu GDE.

FIG. 12 | Schematic diagram of designed flow cell reactor. In the configuration of the flow cell reactor, the Ni foam is positioned in the anode chamber. During the electrochemical measurements, the Ni foam is immersed in the anolyte as the OER catalyst. The oxygen bubbles are removed from the cell via the flow of electrolyte.

FIG. 13 | NMR spectrum of liquid products. Representative ¹H-NMR spectrum of catholyte after CORR on Ag-doped Cu GDE at −0.46 VRHE in 1M KOH. DMSO is used as an internal standard. The peak near 2.21 ppm is assigned to acetone which is used to wash NMR tubes.

FIG. 14 | Electrochemical surface area measurement. Determination of double-layer capacitances over a range of scan rates for different catalysts in 1 M KOH saturated with Ar: a, b, Cu catalyst; c, d, Ag-doped Cu catalyst. The colour scheme in (a) also applied to (c).

FIG. 15 | ECSA-normalized partial n-propanol current densities of Ag-doped Cu and Cu catalysts. Inset, surface roughness factors of catalysts calculated by defining the surface roughness factor for electropolished polycrystalline Cu with an electric double layer capacitance of 29 qF as 1 (ref. 14). The potentials shown here are without iR compensation.

FIG. 16 | The first derivatives of the Cu K-edge XANES spectra of Ag-doped Cu following 10 s at −0.46 VRHE during CORR.

FIG. 17 | SEM images of catalysts. a, Cu—Ag-20 min catalyst. b, Cu—Ag-2 h catalyst.

FIG. 18 | Compositional characterization of a Cu—Ag-20 min nanoparticle. a, HAADF-STEM image of a Cu—Ag-20 min nanoparticle. b-d, The corresponding EELS elemental mappings of Cu, Ag, and O. In Supplementary FIG. 30d, little oxygen signal can be detected in the center of nanoparticle, suggesting that the concentration of Cu oxide in this part is very low. e, Overlap of the corresponding EELS elemental mappings of Cu, Ag, and O.

FIG. 19 | Compositional characterization of a Cu—Ag-2 h nanoparticle. a, HAADF-STEM image of a Cu—Ag-2 h nanoparticle. b-d, The corresponding EELS elemental mappings of Cu, Ag, and O. e, Overlap of the corresponding EELS elemental mappings of Cu, Ag, and O.

FIG. 20 | Structural characterizations of GDE. a,b, WAXS maps for Cu—Ag-20 min GDE (a) and Cu—Ag-2 h GDE (b). c, Sector-averages of WAXS maps for Cu—Ag-20 min and Cu—Ag-2 h GDE in Supplementary FIG. 31a and b. d, XRD patterns for Cu—Ag-20 min and Cu—Ag-2 h GDE.

FIG. 21 | XPS spectra of Cu—Ag-20 min and Cu—Ag-2 h GDE. a, High-resolution Cu 2p spectra of Cu—Ag-20 min and Cu—Ag-2 h GDE. b, High-resolution Ag 3d spectra of Cu—Ag-20 min and Cu—Ag-2 h GDE, demonstrating that the valence of silver in the two samples is 0. Based on the WAXS and XRD results in FIG. 20, both Cu—Ag-20 min and Cu—Ag-2 h GDE contained Cu2O due to the oxidation of Cu during preparation. Cu 2 p XPS results showed that Cu—Ag-20 min GDE contained CuO on the surface, while no CuO could be observed in Cu—Ag-2 h GDE. These results suggest that the increase of Ag concentration in the samples can suppress the further oxidation of Cu2O to CuO.

FIG. 22 | FEs of n-propanol on Ag-doped Cu with different atomic percentages (at %) of Ag and pristine Cu under CORR at a constant potential of −0.46 VRHE.

FIG. 23 | FEs of n-propanol, ethanol, acetate, and ethylene on Ag-doped Cu PTFE electrode during the operation of CORR for 200 min.

FIG. 24 | Structural and compositional analyses of Ag-doped Cu nanoparticles after running CORR for 200 min. a, Low magnification HAADF-STEM image of Ag-doped Cu nanoparticles. b, HAADF-STEM image of an Ag-doped Cu nanoparticle. c-e, The corresponding EELS elemental mappings of Cu, Ag, and O. f, Overlap of the corresponding EELS elemental mappings of Cu, Ag, and O. After the reaction, nanocatalysts were oxidized again during the preparation of the TEM sample in air. Therefore, oxygen was detected in EELS elemental mapping.

FIG. 25 | CO₂ electroreduction performance of Ag-doped Cu and Cu catalyst in flow cell. a, Partial current density of total C2+ products on Ag-doped Cu and pristine Cu under different potentials. b, FEs of acetate, ethylene, n-propanol, and ethanol on Ag-doped Cu and pristine Cu catalysts under different potentials. c, Comparison of FEs and partial current density of n-propanol on Ag-doped Cu and Cu catalysts under different potentials. d, Comparison of _C2+ and n-propanol FEs on Ag-doped Cu and Cu catalysts. All the potentials shown here are without iR compensation.

DETAILED DESCRIPTION

The present description relates to metal catalyst materials to promote the formation of C3 compounds from reactants, such as CO and CO₂ gas, in electroreduction conditions as well as related processes for producing the C3 compounds and for manufacturing the catalyst materials. The present description particularly relates to electroreduction multi-metallic catalysts including a primary metal, such as copper (Cu), and one or more dopant metals, such as silver (Ag) and Ruthenium (Ru). The electroreduction catalysts can be composed so as to have asymmetric active sites, providing a structure to interact with two adsorbates to catalyse C2-C1 coupling, thereby promoting formation of C3 compounds.

Electroreduction of C1 feed gas to high-energy-density fuels provides an attractive avenue to the storage of renewable electricity. Much progress has been made to improve selectivity to C1 and C2 products; however, the selectivity to desirable high-energy-density C3 products remains relatively low. It was reasoned that C3 electrosynthesis relies on a higher-order reaction pathway that requires the formation of multiple carbon-carbon (C—C) bonds; and thus pursued a strategy explicitly designed to couple C2 with C2 intermediates. This work developed an approach to construct asymmetric active sites by doping a very low amount of Ag in Cu via galvanic replacement. The asymmetric active site contains two neighbouring copper atoms with distinct electronic structures, which can interact with two adsorbates to catalyse an asymmetric reaction, thereby boosting C2-C1 coupling. This work achieved a notable Faradaic efficiency (FE) of 33±1% with a conversion rate of 4.5±0.1 mA cm⁻²; and a notable cathodic energy conversion efficiency (EE) of 21%, all for n-propanol. This innovation also represents the first report of CO electroreduction to C3 based on multi-metallic (e.g., bimetallic or trimetallic) catalysts.

The following provides descriptions of some embodiments, implementations, features, aspects, and experimentation that was conducted in the context of the instant technology.

The below description relates to the efficient upgrading of CO to C₃ fuel using asymmetric C-C coupling active sites.

The electroreduction of C₁ feedgas to high-energy-density fuels provides an attractive avenue to the storage of renewable electricity. Much progress has been made to improve selectivity to C₁ and C₂ products; however, the selectivity to desirable high-energy-density C₃ products remains relatively low. This work reasoned that C₃ electrosynthesis relies on a higher-order reaction pathway that requires the formation of multiple carbon-carbon (C-C) bonds; and thus pursue a strategy explicitly designed to couple C₂ with C₁ intermediates. This work developed an approach wherein neighbouring copper atoms having distinct electronic structures interact with two adsorbates to catalyze an asymmetric reaction. This work achieved an n-propanol Faradaic efficiency (FE) of (33±1) % with a conversion rate of (4.5±0.1) mA cm⁻², and an n-propanol cathodic energy conversion efficiency (EE_{cathodic half-cell}) of 21%.

Introduction

The CO₂ electroreduction reaction (CO₂RR) to high-energy-density liquid products is an attractive avenue to achieving the storage of renewable energy. In recent years, much progress has been made in CO₂RR, but the main products reported have been C₁ (CO, CH_4, methanol, and formate) and C₂ (ethylene, acetate, and ethanol) products. N-propanol, a high-value and high-energy-density C₃ product that can be generated via either CO₂ or CO electroreduction, has been produced with low-to-moderate Faradaic efficiencies (FEs) in prior reports.

In CO₂RR, the formation of multi-carbon products starts with the formation of the CO intermediate, followed by the CO electroreduction process. Recently, significant progress has been made in CO₂RR to CO with a FE nearly 100%. To achieve the ultimate goal of high selectivity to high-value-added C₃ products from CO₂RR, it is of interest to improve significantly the FE for CO electroreduction (CORR) to C₃ products.

The formation of C₃ products from CORR relies on the sequential formation of two carbon-carbon (C—C) bonds, the main reaction mechanism for C₃ formation reported previously. Cu provides excellent C-C coupling and produces multi-carbon chemicals in the electroreduction of CO; however, the selectivity towards C₃ products on Cu has remained low. The generation of C₃ products from CO requires multiple product/intermediate formation steps, and it is prone to the competing production of a wide variety of chemical products.

Increasing selectivity in the electroreduction of CO to C₃ products is thus an important challenge to address in the field of electrocatalysis. Until now, catalysts for CORR have focused on Cu and oxide-derived Cu catalysts, and a number of factors have been found to increase performance: these include engineering the oxidation state of the atoms making up the metal catalyst, as well as grain-boundary effects and the selective formation of desired facets. However, the main products of these Cu and oxide-derived Cu catalysts have been C₂ chemicals (ethanol, acetate, and ethylene), and the selectivity to C₃ products has saturated in recent manuscripts based on Cu catalysts.

Here this work explored instead a doping strategy involving different metal-doped Cu (M-doped Cu) catalysts in an attempt to increase C₃ production in CORR. The low C₃ selectivity on Cu catalysts is associated with the low rate of C-C bond formation, including C₁-C₁ and C₁-C₂ coupling. Mechanisms underpinning C₁-C₁ coupling to C₂ products have been explored extensively in prior studies; while C₁-C₂ coupling to C₃ products is less explored. Here this work screened the propensity to catalyze C₁-C₁ and C₁-C₂ coupling using density functional theory (DFT) based on M-doped Cu catalysts. This work found that, among different M-doped Cu candidates explored, Ag-doped Cu can offer the highest activity for both C₁-C₁ and C₁-C₂ coupling, and this work pinpointed a role for the asymmetric C-C coupling active site in this high activity. Specifically, this active site consists of two neighbouring Cu atoms that exhibit different electronic structures: this asymmetry among the neighbours' energetics is determined by the combination of strain and ligand effects arising upon Ag doping. This work then fabricated Ag-doped Cu nanocatalysts via a galvanic replacement approach. The work demonstrated that the synthesized Ag-doped Cu catalyst exhibits higher FEs for n-propanol compared to all previous CORR reports. This leads to superior energy conversion efficiency in the cathodic half-cell (EE_{cathodic half-cell}) for n-propanol.

The nanocatalysts could also include a second dopant metal in addition to Ag to produce a trimetallic catalyst. The second dopant metal can be Ru or another metal suitable for doping along with Ag. The second dopant metal can be incorporated into the material using techniques known in the art; and is selected for compatibility with the Cu and Ag metals as well as the electrocatalytic applications of the catalyst material. The second dopant metal can also be selected to enhance certain properties of the catalyst material and its operation in electrocatalysis applications.

Results

Computational Modelling and Catalyst Design Principles

Catalysts that convert CO into C₃ chemicals require high activity for both C₁-C₁ and C₁-C₂ coupling. With the goal of designing better catalysts for C₃ production, this work investigated the energetics of C₁-C₁ and C₁-C₂ coupling reactions to illustrate the C₂ and C₃ formation rates with the aid of DFT (more details of DFT methods and choice of sequential mechanism can be found in Supplementary Information). Several M-doped Cu systems (M=Ag, Au, Ru, Rh, and Pd) were considered because bimetallic catalysts have been shown to tune catalyst performance in other catalytic reactions. CO dimerization is one reaction pathway for C₁-C₁ coupling, and thus this work used the barrier of CO dimerization to describe the readiness of C₁-C₁ coupling. Due to the abundance of CO species in CORR, the work used the barrier of OCCO and CO coupling as the indicator for the C₁-C₂ coupling (e.g., FIGS. 5-6 as two examples, and Supplementary Tables 1-3). As shown in FIG. 1a, among the M-doped Cu systems studied, calculation results show that Ag-doped Cu bimetallic catalysts that possess the lowest activation energies for both C₁-C₁ and C₁-C₂ coupling, suggesting that doped Cu is a promising catalyst for the formation of C₃ products from CO.

This work carried out further theoretical investigations to uncover physical origins of enhancement in C₁-C₁ and C₁-C₂ coupling on Ag-doped Cu. The work used a model with one Ag atom doped in a Cu(111) slab with a 3×3 unit cell, which corresponds to about 3% doping concentration for four layers of Cu atoms. By a margin of 0.63 eV, Ag at the surface of Cu(111) is more favourable than in subsurface of Cu(111), inducing us to focus on surface-localized Ag in the ensuing studies.

As the radius of the Ag atom is larger than that of Cu atom, Ag doping produces surface strain. The bond length of Cu—Cu changed from 2.57 Å on the Cu(111) surface to 2.55 Å and 2.48 Å on the Ag-doped Cu surface, resulting in asymmetric compressive strain. The ligand effect caused by Ag doping in Cu also has the potential to affect C₁-C₁ and C₁-C₂ coupling. To evaluate the effect of the strain and ligand, a model was built with the same strain of Ag-doped Cu but without Ag substitution by fixing the bond length of Cu surface (denoted Cu with strain). This work calculated the activation energies for C₁-C₁ and C₁-C₂ coupling steps on Cu, Cu with strain, and Ag-doped Cu models (Supplementary Table 4). Calculation results show that both compressive strain and ligand effects contribute to enhanced activity for C₁-C₁ and C₁-C₂ coupling (FIG. 7a).

Ag doping in Cu leads to two classes of neighbouring Cu atoms (denoted Cu-a and Cu-b atoms, FIG. 7b). The two classes exhibit distinct electronic structures, as a result of strain and ligand effects. As to the strain effect, the bond length between Cu-a and Cu-b atoms is 2.55 Å, while the bond length between two Cu-b atoms is 2.48 Å, suggesting that Cu-a is more isolated than Cu-b. As to ligand effects, Cu-a coordinates with nine Cu atoms, while Cu-b coordinates with eight Cu atoms and one Ag atom (FIG. 7). The present disclosure terms a pair of adjacent Cu atoms with different electronic structures an “asymmetric C-C coupling active site”. In this reaction, the asymmetric site interacts with two CO to yield asymmetric reactants—two adsorbed CO on Cu-a and Cu-b atoms with different electronic structures—enhancing C₁-C₁ coupling. In light of the similarity with C—C coupling, the same site can further promote C₁-C₂ coupling between asymmetric C₁ and C₂ intermediates. These findings are in ways analogous to the enhanced coupling effect of Cu⁰ and Cu⁺ proposed by Goddard and co-workers. Taken together, these DFT calculation results suggest that Ag-doped Cu with asymmetric C-C coupling active sites is a good candidate for C₃ production in CORR which appear to support both C₁-C₁ dimerization and C₁-C₂ coupling. It is worth mentioning that there exist other C₁-C₂ coupling possibilities: this work also examined another C₁-C₂ coupling mechanism, OC—OCCOH (refs. 33, 39), and found that Ag-doped Cu had a barrier of 0.76 eV, lower than 0.88 eV on Cu (see Supplementary Table 5).

Preparation and Characterization of Nanocatalysts

This work sought to prepare experimentally Ag-doped Cu catalysts. The work employed a galvanic replacement reaction driven by the difference in the reduction potential of Ag vs. Cu (ref. 46). Firstly, this work deposited a thin layer of commercial Cu nanoparticles with average size of 100 nm on a carbon-based gas diffusion layer (GDL) via spray-coating (FIG. 8a). The Cu gas diffusion electrode (GDE) was then immersed in N₂-saturated 5 □mol L⁻¹ AgNO₃ aqueous solution at 65°C. for 1 h to obtain the Ag-doped Cu GDE. The Ag-doped Cu catalyst retains the particle size and the morphology of the pristine Cu nanoparticles (FIG. 2a, b, and FIGS. 8 and 9). Electron energy loss spectroscopy (EELS) elemental mapping showed that Ag and Cu elements were uniformly distributed in the particle (FIG. 2c and FIG. 10). In transmission wide angle X-ray scattering (WAXS) data and powder X-ray diffraction (XRD) patterns, this work observed peaks for Cu₂O in both Cu and Ag-doped Cu GDE (FIG. 2d and e, and FIG. 11), which were attributed to oxidation of Cu in air during the preparation of GDE. Due to the low concentration of Ag (4.0% in atomic percentage determined by XPS), a peak shift relative to Cu is not observed in WAXS data of the Ag-doped Cu GDE (FIG. 2e and Supplementary Table 6). X-ray photoelectron spectroscopy (XPS) of the Ag-doped Cu GDE also showed the existence of CuO on the surface and confirmed the presence of Ag⁰ (FIG. 2f and g).

CORR Performance and Operando X-ray Absorption Spectroscopy

This work then investigated the performance of the Ag-doped Cu GDE in CORR flow cell reactors (FIG. 12). Flow cells overcome the mass transfer limitation of CO and produce a triple-phase interface that allows the gas reactant to contact the catalyst-electrolyte interface during the reaction. FIG. 3a shows FEs for C₂₊ products in the applied potential range of −0.36 V to −0.56 V with reference to the reversible hydrogen electrode (RHE) in 1 M KOH electrolyte. The liquid products (n-propanol, ethanol, and acetate) and gas products (ethylene and H₂) were quantified using nuclear magnetic resonance (NMR) and gas chromatography, respectively (FIG. 13 and Supplementary Table 7). In the applied potential range of −0.36 V_RHE to −0.56 V_RHE, the total C₂₊ FEs on Ag-doped Cu GDE are higher than that on Cu GDE: indeed the total FE of C₂₊ products on Ag-doped Cu GDE reaches about 80% at −0.56 V_RHE. In particular, at a low potential of −0.46 V_RHE, the Ag-doped Cu GDE records a high n-propanol FE of (33±1) % with the partial n-propanol current density of (4.5±0.1) mA cm⁻², whereas n-propanol FE on pristine Cu is (22±1)% (FIG. 3b). This impressive n-propanol FE represents the highest value reported for n-propanol production via CO₂RR and CORR (Supplementary Table 8). The higher FE for C₂₊ and C₃ products on Ag-doped Cu relative to Cu is consistent with predictions from DFT. The intrinsic activities for n-propanol production on Ag-doped Cu and Cu are reported via the partial current density for n-propanol production normalized to the electrochemical surface area (ECSA) (FIGS. 14, 15, and Supplementary Table 9). The ECSA-normalized partial n-propanol current density on Ag-doped Cu is 0.124 mA cm⁻², which is 3 times that on Cu. The n-propanol EE_{cathodic half-cell} reaches 20% at a low potential of −0.46 V_RHE when the overpotential of oxygen evolution in anode side is assumed to be 0. After correcting for ohmic loss (Supplementary Table 10), the n-propanol EE_{cathodic half-cell} reaches 21% under a low overpotential of 0.616 V. This EE_{cathodic half-cell} is higher than the best prior reports by a margin of 1.3× (Supplementary section, and Supplementary Table 8).

It should be noted that the highest n-propanol FE on both Ag-doped Cu and Cu GDEs were achieved at relatively low potential (−0.46 VRHE), and n-propanol FE decreased when further increasing the potential to −0.56 V_RHE (FIG. 3a and Supplementary Table 7). The total C₂ product and ethylene FEs on both Ag-doped Cu and Cu GDEs exhibited an increasing trend with increased applied potentials. This result can be explained by noting that the C—C coupling step for n-propanol formation becomes slow at high potential, and thus C₂ intermediate protonation reaction is more favored compared with C—C coupling step for n-propanol formation.

To study the chemical state of the catalyst during CORR, this work performed operando X-ray absorption spectroscopy (XAS) of Ag-doped Cu GDE under operando CORR conditions at a constant applied potential of −0.46 VRHE. Operando Cu K-edge X-ray absorption near edge structure (XANES) spectra of Ag-doped Cu GDE show that Cu atoms were reduced to Cu⁰ in the first 10 s during CORR (FIG. 3c and FIG. 16). Thereafter, the valence state of Cu is maintained at zero throughout CORR. Consistent results are shown in the operando extended X-ray adsorption fine structure (EXAFS). An EXAFS fitting analysis at the Cu K-edge showed the presence of Ag and its strong interaction with Cu by having Cu—Ag distance value between pure Cu and pure Ag (FIG. 3d and Supplementary Table 11). Collectively, operando XAS results demonstrated that Ag atoms are doped in the lattice of Cu nanoparticles, and that each element remains in its metallic state during CORR. As a control experiment, this work carried out operando Cu K-edge XANES on Cu GDE, and it also showed that Cu oxides were reduced to Cu⁰ in the first 10 s during CORR. These results further demonstrated that, rather than benefiting from oxidation states, the high FE of n-propanol on Ag-doped Cu GDE is associated with metallic states of Cu and might be ascribed to the different structures of Ag-doped Cu relative to Cu.

To explore further the role of Ag doping in Cu in facilitating C₁-C₁ and C₁-C₂ coupling, this work investigated the CORR performance of Ag-doped Cu catalysts with different Ag concentrations. To vary the Ag concentration, this work changed the immersion time of the Cu GDE in AgNO3 solution to 20 min and 2 h, respectively, denoted Cu—Ag-20 min and Cu—Ag-2 h. As in Ag-doped Cu, both Cu—Ag-20 min and Cu—Ag-2 h catalysts also retained the particle size and the morphology of the pristine Cu nanoparticles (FIGS. 17-19). Thus, the size and morphology effect can be excluded when comparing the CORR performance between Cu and different types of Ag-doped Cu catalysts. The atomic percentages of Ag doping in Cu were tuned to 2.4% and 7.8% (XPS), respectively, for Cu—Ag-20 min and Cu—Ag-2 h catalysts (FIGS. 20, 21, and Supplementary Table 6). Relative to Cu GDE, both Cu—Ag-20 min and Cu—Ag-2 h GDE exhibited enhanced FEs of n-propanol, as well as higher FEs of total C₂₊products (FIGS. 4a and b), in the potential range from −0.36 VRHE to −0.56 VRHE toward CORR. These results are in agreement with DFT predictions of enhancement in C₁-C₁ and C₁-C₂ coupling by Ag doping in Cu. Additionally, among Ag-doped Cu catalysts with different Ag concentrations, Ag-doped Cu with an atomic percentage of 4.0% Ag showed the highest n-propanol FE at −0.46 VRHE (FIG. 22). The analysis of operando Cu K-edge XANES of Cu—Ag-20 min and Cu—Ag-2 h GDE also demonstrates that Cu oxides are reduced to Cu⁰ in the first 10 s during CORR (FIG. 4c) and it is Ag-doped Cu in metallic state that is associated with the enhanced selectivity to n-propanol during CORR.

The carbon-based gas diffusion layers suffer from liquid penetration and gas diffusion blockage, termed flooding, over time. To overcome the flooding issue on carbon-based GDE after long operation time, this work fabricated the Ag-doped Cu polytetrafluoroethylene (PTFE) electrode based on a configuration (graphite/carbon nanoparticle/Ag-doped Cu/PTFE electrode) that has been developed. The Ag-doped Cu layer was prepared by immersing a Cu layer in 5 □mol L⁻¹ AgNO₃ aqueous solution at 65° C. for 1 h. When a potential of −0.46 V_RHE was applied, the FE of n-propanol on the Ag-doped Cu PTFE electrode achieved 33% and operated stably over 200 min of CORR (FIGS. 23 and 24).

This work also compared the CO₂RR performance of Ag-doped Cu and pristine Cu to explore whether the asymmetric active sites enhance in this distinct context the C₁-C₁ and C₁-C₂ coupling (FIG. 25 and Supplementary Table 12). In the potential range from −0.82 VRHE to −1.79 VRHE (after iR compensation), both C₂₊ and C₃ FEs on Ag-doped Cu catalysts are notably higher than those on pristine Cu: at the potential of −2.96 V_RHE (−1.31 V_RHE after iR compensation), the partial C2+ and n-propanol current densities of Ag-doped Cu are 308±6 mA cm⁻² and 36±2 mA cm⁻², and C₂₊ and n-propanol FEs on Ag-doped Cu are 62% and 7%, respectively, providing a doubling compared to pristine Cu.

Discussion

This work demonstrates Ag doping in Cu to facilitate C₁-C₁ and C₁-C₂ coupling and thus improve the selectivity to C₃ products during CORR. DFT results show that the strain and ligand effects due to Ag doping jointly provide an asymmetric C—C coupling active site containing two neighbouring Cu atoms with different electronic structures, and that these are capable of enhancing C₁-C₁ and C₁-C₂ coupling. Experimentally, this work achieved a total C₂₊ FE of about 80% and a record n-propanol FE of (33±1) % with a partial n-propanol current density (4.5±0.1) mA cm⁻² on Ag-doped Cu catalyst in CORR. The EE_{cathodic half-cell} for n-propanol also reaches 21% at a low potential of 0.416 V_RHE, with a low overpotential of 0.616 V. These findings provide a framework for rational catalyst design for tuning the CORR selectivity towards high-energy-density C₃ liquid products, a relevant step in overcoming the bottleneck of the electroproduction of C₃ products.

Methods

DFT Calculations

In this work, all the DFT calculations were carried out with a periodic slab model using the Vienna ab initio simulation program (VASP) (https://www.vasp.at/). Detailed theoretical methods are found in the Supplementary Information.

Preparation of Electrodes

8.5 mg of commercial Cu was dispersed in a mixture of 0.85 mL of methanol and 8.5 □l of 5% Nafion under ultrasonication for 30 min. The suspension was deposited on a carbon-based GDL using spray-coating with a catalyst loading of ≈1 mg cm⁻² to prepare the Cu GDE. The prepared Cu GDE was immersed in 5 □mol L⁻¹ AgNO₃ aqueous solution at 65° C. for a certain time period to prepare Ag-doped Cu GDE as cathodes. The main goal of the work was to focus on improving the efficiency of the cathodic side of CORR to propanol. Thus, the work used Ni foam (1.6 mm thickness, MTI Corporation) as the oxygen evolution reaction (OER) catalyst in the anode side because it is commercially available and have been showed as a good OER catalyst. Details of chemicals and materials information are found in the Supplementary Information.

Characterization

Scanning electron microscopy (SEM) images were taken using a Quanta FEG 250 microscope. (Certain commercial equipment, instruments, or materials are identified in this paper and Supplementary Information in order to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by the National Institute of Standards and Technology, nor is it intended to imply that the materials or equipment identified are necessarily the best available for the purpose). HAADF-STEM images were taken using an aberration-corrected FEI Titan 80-300 kV TEM/STEM microscope at 300 kV, with a probe convergence angle of 30 mrad and a large inner collection angle of 65 mrad to provide a nominal image solution of 0.7 Å. EELS elemental mapping was collected on aberration-corrected JEOL JEM-ARM200F electron microscope at 200 kV equipped with Gatan GIF quantum energy filters. Structural characterization of cathodes was obtained using XRD (MiniFlex600) with Cu-K□ radiation. The surface compositions of cathodes were determined by XPS (model 5600, Perkin-Elmer) using a monochromatic aluminum X-ray source. Operando XAS measurement were conducted at 9BM beamline at Advanced Photon Source (APS, Argonne national laboratory, IL). Athena and Artemis software included in a standard IFEFFIT package were used to process XAS data. WAXS measurements were carried out in transmission geometry at the CMS beamline of the National Synchrotron Light Source II (NSLS-II), a U.S. Department of Energy (DOE) office of the Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory. Samples were measured with an imaging detector at a distance of 0.177 m using X-ray wavelength of 0.729 Å. Nika software package was used to sector average the 2D WAXS images. Data plotting was done in Igor Pro (Wavemetrics, Inc., Lake Oswego, Oreg., USA).

Electrochemical Measurements

All the electrochemical measurements were conducted in flow cell reactor. Electrocatalytic measurements were operated using the three-electrode system at an electrochemical station (AUT50783). In the flow cell reactor, the prepared GDEs, anion exchange membrane, and nickel foam were positioned and clamped together between silicone gaskets and PTFE flow fields. Then 10 mL of electrolyte (1 M KOH aqueous solution) was introduced into the anode chamber between anode and membrane, as well as the cathode chamber between membrane and cathode, respectively. The electrolytes in cathode and anode were circulated by two pumps at the rate of 10 mL min⁻¹. CO gas (Linde, 99.99%) or CO₂ gas (Linde, 99.99%) was continuously supplied to gas chamber located at the back side of cathode GDE at the rate of 50 mL min⁻¹. Gas could diffuse into the interface between cathode and electrolyte, thus generating a triple-phase interface between gas, electrode, and electrolyte. The catalytic performance of cathodes was evaluated by performing potentiostatic electrolysis.

All potentials were measured against an Ag/AgCl reference electrode (3 M KCl, BASi). Gas and liquid products were respectively analyzed using gas chromatograph (PerkinElmer Clarus 600) equipped with thermal conductivity and flame ionization detectors, and NMR spectrometer (Agilent DD2 600 MHz) by taking dimethylsulfoxide (DMSO) as an internal standard. All the potentials were converted to values with reference to RHE using: E_RHE=E_Ag/AgCl+0.210 V+0.0591×pH.

ECSA was determined based on the equation: ECSA=R_fS, where R_f was roughness factor and S was geometric area of electrode (1 cm⁻²). R_f=C_dI/29 □F cm⁻², where C_dl is the double layer capacitance of catalyst and the double-layer capacitance of a smooth Cu surface is assumed to be 29 □F cm⁻² (ref. 21). Double layer capacitances of catalysts were determined by measuring cyclic voltammetry with different scan rates (40, 60, 80, 100, 120, and 140 mV s⁻¹, respectively) in the potential ranges between 0.20 V_RHE and 0.24 V_RHE where no Faradaic process occurred. The cyclic voltammetry (CV) measurement was operated in the same flow cell reactor and 1 M KOH aqueous solution saturated with nitrogen (Linde, 99.998%) was used as the electrolyte. The flow cell reactor was filled with electrolyte prior to the CV measurement and the electrolyte was not circulated during the CV measurement. N₂, instead of CO₂, was continuously supplied to gas chamber of the cell. By plotting the average current j (j=(j_a−j_c)/2, where j_a and j_c are anodic and cathodic current densities, respectively) at 0.22 V_RHE against the scan rate, C_dl value was given by the slope.

Electrochemical impedance spectroscopy (EIS) technique was used to measure the ohmic loss between the working and reference electrodes and 70% iR compensation was applied to correct the potentials manually.

References. The following references and all others mentioned herein are incorporated herein by reference.

1. Liu, M. et al. Enhanced electrocatalytic CO₂ reduction via field-induced reagent concentration. Nature 537, 382-386 (2016).

2. Gao, S. et al. Partially oxidized atomic cobalt layers for carbon dioxide electroreduction to liquid fuel. Nature 529, 68-71 (2016).

3. Seifitokaldani, A. et al. Hydronium-induced switching between CO₂ electroreduction pathways. J. Am. Chem. Soc. 140, 3833-3837 (2018).

4. Le, M. et al. Electrochemical reduction of CO₂ to CH₃OH at copper oxide surfaces. J. Electrochem. Soc. 158, E45-E49 (2011).

5. Zhang, W. et al. Electrochemical reduction of carbon dioxide to methanol on hierarchical Pd/SnO₂ nanosheets with abundant Pd—O—Sn interfaces. Angew. Chem. Int. Ed. 57, 9475-9479 (2018)

6. Qiao, J., Liu, Y., Hong, F. & Zhang, J. A review of catalysts for the electroreduction of carbon dioxide to produce low-carbon fuels. Chem. Soc. Rev. 43, 631-675 (2014).

7. Dinh, C.-T. et al. CO₂ electroreduction to ethylene via hydroxide-mediated copper catalysis at an abrupt interface. Science 360, 783-787 (2018).

8. Mistry, H. et al. Highly selective plasma-activated copper catalysts for carbon dioxide reduction to ethylene. Nat. Commun. 7, 12123 (2016).

9. Liu, Y., Chen, S., Quan, X. & Yu, H. Efficient electrochemical reduction of carbon dioxide to acetate on nitrogen-doped nanodiamond. J. Am. Chem. Soc. 137, 11631-11636 (2015).

10. Zhuang, T.-T. et al. Steering post-C—C coupling selectivity enables high efficiency electroreduction of carbon dioxide to multi-carbon alcohols. Nat. Catal. 1, 421-428 (2018).

11. Ren, D., Ang, B. S.-H. & Yeo, B. S. Tuning the selectivity of carbon dioxide electroreduction toward ethanol on oxide-derived Cu_xZn catalysts. ACS Catal. 6, 8239-8247 (2016).

12. Kim, D., Kley, C. S., Li, Y. & Yang, P. Copper nanoparticle ensembles for selective electroreduction of CO₂ to C₂-C₃ products. Proc. Natl. Acad. Sci. USA 114, 10560-10565 (2017).

13. Hoang, T. T. H. et al. Nanoporous Copper-silver alloys by additive-controlled electrodeposition for the selective electroreduction of CO₂ to ethylene and ethanol. J. Am. Chem. Soc. 140, 5791-5797 (2018).

14. Jouny, M., Luc, W. & Jiao, F. General techno-economic analysis of CO₂ electrolysis systems. Ind. Eng. Chem. Res. 57, 2165-2177 (2018).

15. Clark, E. L., Hahn, C., Jaramillo, T. F. & Bell, A. T. Electrochemical CO₂ reduction over compressively strained CuAg surface alloys with enhanced multi-carbon oxygenate selectivity. J. Am. Chem. Soc. 139, 15848-15857 (2017).

16. Verdaguer-Casadevall, A. et al. Probing the active surface sites for CO reduction on oxide-derived copper electrocatalysts. J. Am. Chem. Soc. 137, 9808-9811 (2015).

17. Handoko, A. D. et al. Mechanistic insights into the selective electroreduction of carbon dioxide to ethylene on Cu₂O-derived copper catalysts. J. Phys. Chem. C 120, 20058-20067 (2016).

18. Ren, D. et al. Mechanistic insights into the enhanced activity and stability of agglomerated Cu nanocrystals for the electrochemical reduction of carbon dioxide to n-propanol. J. Phys. Chem. Lett. 7, 20-24 (2016).

19. Rahaman, M., Dutta, A., Zanetti, A. & Broekmann, P. Electrochemical reduction of CO₂ into multicarbon alcohols on activated Cu mesh catalysts: an identical location (IL) study. ACS Catal. 7, 7946-7956 (2017).

20. Jiang, K. et al. Metal ion cycling of Cu foil for selective C—C coupling in electrochemical CO₂ reduction. Nat. Catal. 1, 111-119 (2018).

21. Li, C. W., Ciston, J. & Kanan, M. W. Electroreduction of carbon monoxide to liquid fuel on oxide-derived nanocrystalline copper. Nature 508, 504-507 (2014).

22. Feng, X., Jiang, K., Fan, S. & Kanan, M. W. A direct grain-boundary-activity correlation for CO electroreduction on Cu nanoparticles. ACS Cent. Sci. 2, 169-174 (2016).

23. Raciti, D. et al. Low-overpotential electroreduction of carbon monoxide using copper nanowires. ACS Catal. 7, 4467-4472 (2017).

24. Li, J. et al. Copper adparticle enabled selective electrosynthesis of n-propanol. Nat. Commun. 9, 4614 (2018).

25. Zhuang, T.-T. et al. Copper nanocavities confine intermediates for efficient electrosynthesis of C3 alcohol fuels from carbon monoxide. Nat. Catal. 1, 946-951 (2018)

26. Jouny, M., Luc, W. & Jiao, F. High-rate electroreduction of carbon monoxide to multi-carbon products. Nat. Catal. 1, 748-755 (2018).

27. Kortlever, R., Shen, J., Schouten, K. J. P., Calle-Vallejo, F. & Koper, M. T. M. Catalysts and reaction pathways for the electrochemical reduction of carbon dioxide. J. Phys. Chem. Lett. 6, 4073-4082 (2015).

28. Cheng, T., Xiao, H. & Goddard, W. A. Full atomistic reaction mechanism with kinetics for CO reduction on Cu(100) from ab initio molecular dynamics free-energy calculations at 298 K. Proc. Natl. Acad. Sci. USA 114, 1795-1800 (2017).

29. Montoya, J. H., Shi, C., Chan, K. & Nørskov, J. K. Theoretical insights into a CO dimerization mechanism in CO₂. J. Phys. Chem. Lett. 6, 2032-2037 (2015).

30. Liu, S. et al. Shape-dependent electrocatalytic reduction of CO₂ to CO on triangular silver nanoplates. J. Am. Chem. Soc. 139, 2160-2163 (2017).

31. Jhong, H.-R. M. et al. A nitrogen-doped carbon catalyst for electrochemical CO₂ conversion to CO with high selectivity and current density. ChemSusChem 10, 1094-1099 (2017).

32. Gabardo, C. M. et al. Combined high alkalinity and pressurization enable efficient CO₂ electroreduction to CO. Energy Environ. Sci. 11, 2531-2539 (2018).

33. Xiao, H., Cheng, T. & Goddard, W. A. Atomistic mechanisms underlying selectivities in C₁ and C₂ products from electrochemical reduction of CO on Cu (111). J. Am. Chem. Soc. 139, 130-136 (2016).

34. Kuhl, K. P., Cave, E. R., Abram, D. N. & Jaramillo, T. F. New insights into the electrochemical reduction of carbon dioxide on metallic copper surfaces. Energy Environ. Sci. 5, 7050-7059 (2012).

35. Vayenas, C. G., White, R. E. & Gamboa-Aldeco, M. E. Modern Aspects of Electrochemistry Vo. 42 (Springer, New York, 2008)

36. Hori, Y., Takahashi, R., Yoshinami, Y. & Murata, A. Electrochemical reduction of CO at a copper electrode. J. Phys. Chem. B 101, 7075-7081 (1997).

37. Loiudice, A. et al. Tailoring copper nanocrystals towards C₂ products in electrochemical CO₂ reduction. Angew. Chem. Int. Ed. 55, 2016, 5789-5792.

38. Lum, Y., Cheng, T., Goddard, W. A. & Ager, J. W. Electrochemical CO reduction builds solvent water into oxygenate products. J. Am. Chem. Soc. 140, 2018, 9337-9340.

39. Pérez-Gallent, E., Figueiredo, M. C., Calle-Vallejo, F. & Koper, M. T. M. Spectroscopic observation of a hydrogenated CO dimer intermediate during CO reduction on Cu(100) electrodes. Angew. Chem. Int. Ed. 56, 3621-3624 (2017).

40. Goodpaster, J. D., Bell, A. T. & Head-Gordon, M. Identification of possible pathways for C—C bond formation during electrochemical reduction of CO₂: new theoretical insights from an improved electrochemical model. J. Phys. Chem. Lett. 7, 1471-1477 (2016).

41. Xiao, H., Goddard, W. A., Cheng T. & Liu, Y. Cu metal embedded in oxidized matrix catalyst to promote CO₂ activation and CO dimerization for electrochemical reduction of CO₂. Proc. Natl. Acad. Sci. USA 114, 6685-6688 (2017).

42. Hussain, J., Jonsson, H. & Skulason, E. Calculations of product selectivity in electrochemical CO₂ reduction. ACS Catal. 8, 5240-5249 (2018).

43. Greeley, J. et al. Alloys of platinum and early transition metals as oxygen reduction electrocatalysts. Nat. Chem. 1, 552-556 (2009).

44. Studt, F. et al. Identification of non-precious metal alloy catalysts for selective hydrogenation of acetylene. Science 320, 1320-1322 (2008).

45. Liu, P. & Norskov, J. K. Ligand and ensemble effects in adsorption on alloy surfaces. Phys. Chem. Chem. Phys. 3, 3814-3818 (2001).

46. Cobley, C. M. & Xia, Y. Engineering the properties of metal nanostructures via galvanic replacement reactions. Mater. Sci. Eng. R Rep. 70, 44-62 (2010).

47. Zhou, Y. et al. Dopant-induced electron localization drives CO₂ reduction to C₂ hydrocarbons. Nat. Chem. 19, 974-980 (2018).

48. Zhu, W., Zhang, R., Qu, F., Asiri, A. M. & Sun, X. Design and application of foams for electrocatalysis. Chemcatchem 9, 1721-1743 (2017).

49. Ravel, B. & Newville, M. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat. 12, 537-541 (2005).

50. Jan, I. Nika: software for two-dimensional data reduction. J. Appl. Crystallogr. 45, 324-328 (2012).

SUPPLEMENTARY TABLE 1
Reaction barriers of C1-C1 and C1-C2 coupling on Cu and
M-doped Cu surfaces.
	Ea (C1-C1)	Ea (C1-C2)
	(eV)	(eV)
Cu	0.72	0.65
Ag-doped Cu	0.69	0.61
Au-doped Cu	0.70	0.64
Pd-doped Cu	0.80	0.75
Ru-doped Cu	0.74	0.61
Rh-doped Cu	0.79	0.62

SUPPLEMENTARY TABLE 2
CO adsorption energies (Ead) on different sites of Ag-
doped Cu.
	Initial	Optimized
	sites	sites	Ead (eV)
	a1	a	−0.37
	b	b	−0.84
	c	c	−0.82
	d	b	−0.80
	e	e	−0.92
	f	n	−0.97
	g	o	−0.96
	h	n	−0.97
	i	n	−0.97
	j	j	−0.96
	k	k	−0.96
	l	l	−0.96
	m	b	−0.84
	n	n	−0.97
	o	o	−0.96
	p	k	−0.96

SUPPLEMENTARY TABLE 3
Activation energies of C1-C1 and C1-C2 coupling on Cu and
Ag-doped Cu with different exchange-correlation functionals.
	Ea (C1-C1)/eV	Ea (C1-C2)/eV
	Cu	AgCu	Cu	AgCu
PBE	0.72	0.69	0.65	0.61
BEFF	0.63	0.58	1.36	0.59
PBEsol	0.80	0.62	0.79	0.52
PW91	0.68	0.63	0.68	0.54

SUPPLEMENTARY TABLE 4
Strain and ligand effect on reaction barriers of C1-C1 and
C1-C2 coupling on Cu, Cu with strain, and Ag-doped Cu.
	Ea (C1-C1)	Ea (C1-C2)
	(eV)	(eV)
Cu	0.72	0.63
Cu with strain	0.72	0.58
Ag-doped Cu	0.69	0.56

SUPPLEMENTARY TABLE 5
Reaction barriers and enthalpy changes of CO—OCCCOH
on Cu and Ag-doped Cu surfaces.
		Ea (eV)	ΔH (eV)
	Cu	0.88	0.05
	Ag-doped Cu	0.76	−0.08

At the low applied potential, the calculated barrier of OC—OCCOH on Cu(111) is 0.88 eV (E_OC—OCCOH), higher than of that of OC—OCCO (0.63 eV, E_OC—OCCO), indicating that the OC—OCCO is more favorable than OC—OCCOH (Supplementary Table 5).

According to the energetic span, the total barriers of OC—OCCO (E₁) and OC—OCCOH (E₂) pathways are:

E₁=ΔH_OC—CO+E_OC—OCCO   (equation 1)

E₂=max(ΔH_OC—CO+ΔH_OCCOH+E_OC—OCCOH+eU,E_OC—OCCOH)   (equation 2),

where U is the applied potential vs. the computational hydrogen electrode (CHE) (ref. 2). The reaction energy of CO dimerization is 0.65 eV (ΔH_OC—CO), and the reaction energy of OCCO hydrogenation is −0.05 eV (ΔH_OCCOH) Therefore, E₂ decreases with the increase of the applied potential. As the reaction potential showed the highest FE_n-propanol in this experiment is −0.46 V_RHE (equal to −1.286 V_CHE), the total barrier is 0.88 eV, giving the maximum turnover frequency (TOF) of 0.01 s⁻¹.

The total barrier of OC−OCCOH on Ag-doped Cu is 0.76 eV, lower than that on pure Cu, giving a maximum TOF of 0.87 s⁻¹ at reaction applied potential (−1.286 V_CHE). Therefore, after considering the proton/electron transfer in C₁-C₂ coupling, the designed Ag-doped Cu also favors C₁-C₂ coupling reaction compared to Cu.

SUPPLEMENTARY TABLE 6
Ag concentrations in different types of Ag-doped Cu GDE
determined using XPS.
	Ag-doped Cu	Cu—Ag-20 min	Cu—Ag-2 h
Atomic percentage	4.0	2.4	7.8
of Ag (%)

SUPPLEMENTARY TABLE 7
Product FEs for different nanocatalysts under different
applied potentials in 1M KOH electrolyte in CORR.
	Potentials1	FEn-propanol	FEethanol	FEacetate	FEethylene	FEhydrogen	FEtotal
Catalysts	(VRHE)	(%)	(%)	(%)	(%)	(%)	(%)
Cu	−0.36	20.2 ± 0.3	12.5 ± 1	7.4 ± 1	17.1 ± 1.1	18.0 ± 3	~75.2
	−0.46	22.4 ± 0.9	9.0 ± 1	5.8 ± 0.9	26.5 ± 1	28.3 ± 4	~92
	−0.56	17.1 ± 0.8	5.0 ± 1.1	6.1 ± 0.6	32.2 ± 1	18.2 ± 4	~78.6
Ag-doped Cu	−0.36	23.6 ± 0.5	7.4 ± 0.7	9.6 ± 0.6	18.0 ± 2	25.2 ± 2	~83.8
	−0.46	33.3 ± 1	5.9 ± 2	4.7 ± 0.8	29.7 ± 3	23 ± 3	~96.6
	−0.56	23.8 ± 1.1	8.1 ± 1.9	4.4 ± 0.4	43.9 ± 1.8	20.2 ± 2	~100.4
Cu-Ag-20 min	−0.36	28.6 ± 0.7	16.1 ± 1.2	16.0 ± 1	14.4 ± 0.9	9.0 ± 2	~83.1
	−0.46	26.8 ± 0.5	13.4 ± 1	4.9 ± 1	34.1 ± 2	20.6 ± 3	~99.8
	−0.56	21.7 ± 1.2	6.5 ± 0.4	4.5 ± 0.4	32.4 ± 3	29.5 ± 1	~94.6
Cu-Ag-2 h	−0.36	31.2 ± 0.5	6.1 ± 0.6	8.6 ± 1	15.1 ± 1.1	8.2 ± 2	~69.2
	−0.46	30.5 ± 0.3	8.4 ± 1	7.2 ± 1.3	26.5 ± 1.7	17.1 ± 1.1	~89.7
	−0.56	24.3 ± 1	9.1 ± 0.5	9.4 ± 0.2	34.7 ± 2	12.6 ± 3	~90.1
1Potentials shown here are without iR compensation.
Error bars in this Table represent the standard deviation based on three separate measurements.

SUPPLEMENTARY TABLE 8
Comparison of n-propanol production via
electroreduction CO2 or CO on Cu-based catalysts.
			N-propanol	N-propanol
	Potentials	Overpotentials	FEs	EEcathodic half-cell
Catalysts	(VRHE)	(mV)	(%)	(%)	Reactions	References
Ag-doped Cu	−0.416a	616	33 ± 1	20.7 ± 0.6	CORR	This work
Cu-Ag-2 h	−0.34a	540	31 ± 0.5	20.3 ± 0.3	CORR	This work
Oxide-derived Cu	—	—	5.4	—	CO2RR	J. Phys. Chem. C 120,
						20058-20067 (2016)
						(Ref. 15)
Agglomerated Cu	−0.95	1050	10.6	5.5	CO2RR	J. Phys. Chem. Lett. 7,
nanocrystals						20-24 (2016). (Ref. 16)
Cu nanoparticles	−0.81	910	5.9	3.0	CO2RR	Proc. Natl. Acad. Sci.
						U.S.A. 114, 10560-
						10565 (2017) (Ref. 17)
Activated Cu mesh	−0.9	1000	13.1	6.9	CO2RR	ACS Catal. 1, 7946-
						7956 (2017) (Ref. 18)
Metal ion cycled Cu	−0.96	1060	15	7.7	CO2RR	Nat. Catal. 1, 111-119
						(2018) (Ref. 19)
Cu2S-Cu-V	−0.95	1050	8 + 0.7	4.9 + 0.4	CO2RR	Nat. Catal. 1, 421-428
						(2018) (Ref. 20)
Oxide-derived Cu	−0.4	600	10.0	6.3	CORR	Nature 508, 504-507
						(2014) (Ref. 14)
Oxide-derived Cu	−0.4	600	8.8	5.6	CORR	J. Am. Chem. Soc. 137,
						9808-9811 (2015) (Ref.
						21)
Cu nanoparticles	−0.5	700	≈6	3.6	CORR	ACS Cent. Sci. 2, 169-
						174 (2016) (Ref. 22)
Cu nanowires	−0.45	650	1.8	1.1	CORR	ACS Catal. 7,4467-
						4472 (2017) (Ref. 23)
Oxide-derived Cu	−0.42	620	25.6	16.0	CORR	Nat. Catal. 1, 748-755
						(2018) (Ref. 24)
Cu adparticles	−0.47	670	23	13.9	CORR	Nat. Commun. 9, 4614
						(2018) (Ref. 25)
Cavity Cu	−0.56	760	21 ± 1	12.0 ± 0.6	CORR	Nat. Catal. 1, 946-951
						(2018) (Ref. 26)
aPotential corrected by ohmic loss.
Error bars in this Table represent the standard deviation based on three separate measurements.

SUPPLEMENTARY TABLE 9
ECSA-normalized partial n-propanol current densities of
Ag-doped Cu catalyst.
Potentials (after iR ECSA-normalized
compensation, VRHE)  jn-propanol (mA cm−2)
−0.344               0.03
−0.416               0.12
−0.449               0.22

SUPPLEMENTARY TABLE 10
Uncompensated resistances of the electrochemical cells
for Ag-doped Cu, Cu—Ag-2 h, and Cu GDE measured by EIS.
	Ag-doped Cu	Cu—Ag-2 h	Cu
Resistances (Ω)	4.72	4.66	5.33

SUPPLEMENTARY TABLE 11
EXAFS fitting parameters of Ag-doped Cu catalyst and
Cu foil.
				σ2 × 1.03	ΔE0 (eV),
Samples	Shells	N	R(Å)	(Å2)	Cu K-edge
Cu foil	Cu—Cu	12	2.545(1)	8.1(5)	5.2(6)
Ag-doped Cu	Cu—Cu	9.9(3)	2.538(4)	7.1(4)	3.0(9)
	Cu—Ag	0.9(2)	2.629(7)
N, coordination number;
R, bonding distance;
σ2, Debye-Waller factor;
ΔE0, shift in adsorption edge energy.

SUPPLEMENTARY TABLE 12
Product FEs for Ag-doped Cu and pristine Cu
nanocatalysts under different applied potentials in 1M KOH electrolyte in CO2RR.
	Potentials1	Potentials2	FEn-propanol	FEethanol	FEacetate	FEethylene	FEformate	FECO	FEhydrogen	FEtotal
Catalysts	(VRHE)	(VRHE)	(%)	(%)	(%)	(%)	(%)	(%)	(%)	(%)
Ag-	−1.96	−0.84	6.8 ± 0.5	10 ± 0.5	0.5 ± 0.1	30.8 ± 0.5	6.8 ± 0.3	35.2 ± 0.5	13.6 ± 0.2	~103.7
doped	−2.96	−1.31	7.1 ± 0.3	12.8 ± 0.3	0.6 ± 0.1	41.1 ± 0.5	4.7 ± 0.3	24.5 ± 0.4	12 ± 0.3	~102.8
Cu	−3.46	−1.50	8.0 ± 0.3	12 ± 0.6	0.5 ± 0.1	39.5 ± 0.2	6.5 ± 0.3	25.5 ± 0.5	11.4 ± 0.7	~103.4
Cu	−1.96	−0.82	3.2 ± 0.1	4.8 ± 0.3	0.5 ± 0.1	14.7 ± 0.5	8.9 ± 0.3	60 ± 1	11.7 ± 0.5	~103.8
	−2.96	−1.68	3.8 ± 0.3	5.8 ± 0.3	0.5 ± 0.1	22.5 ± 0.5	7.6 ± 0.5	50.5 ± 0.4	11.5 ± 0.8	~102.2
	−3.46	−1.79	4.8 ± 0.2	10.8 ± 0.1	0.5 ± 0.1	35.0 ± 0.5	3.8 ± 0.1	31 ± 0.5	11.0 ± 0.5	~96.9
1Potentials before iR compensation.
2Potentials after iR compensation.
Error bars in this Table represent the standard deviation based on three separate measurements.

Supplementary Methods

Theoretical Methods. In this work, all the DFT calculations were carried out with a periodic slab model using the Vienna ab initio simulation program (VASP). The generalized gradient approximation (GGA) was used with the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional. The projector-augmented wave (PAVV) method was utilized to describe the electron-ion interactions, and the cut-off energy for the plane-wave basis set was 450 eV. In order to illustrate the long-range dispersion interactions between the adsorbates and catalysts, this work employed the D3 correction method of Grimme et al. (ref. 10). Brillouin zone integration was accomplished using a ×3×1 Monkhorst-Pack k-point mesh. All the adsorption geometries were optimized using a force-based conjugate gradient algorithm, while transition states (TSs) were located with a constrained minimisation technique. At all intermediate and transition states, one charged layer of water molecules was added to the surface to take the combined field and solvation effects into account. In the CO dimerization, there is no proton or electron transfer, thus the computational hydrogen electrode was not used in this work. For the modelling of Cu(111), the crystal structure was optimized, and Cu(111) was modelled with a periodic four-layer p(4×4) model with the 2 lower layers fixed and 2 upper layers relaxed. Cu(111) was chosen because Cu(111) is more stable relative to Cu(100) (ref. 14), and thus improving the activity of Cu(111) for C₃ formation is more significant. In addition, the overall barrier of C₁ to C₃ product on Cu(100) is higher than that on Cu(111), despite the low barrier of C₁-C₁ dimerization on Cu(100).

This work started with CO adsorption on Cu(111) surface. The reaction barrier of CO dimerization is calculated to be 0.72 eV, which is similar to the value proposed by Nørskov and co-workers in the same solvent model. The barrier of CO and OCCO coupling is calculated to be 0.63 eV. Given that C₂ FEs are always higher than C₃ FEs in previous CORR and CO₂RR reported, C—C coupling is likely more favorable than concerted C—C—C coupling. Thus, rather than concerted C—C—C coupling, sequential formation of C—C bonds was considered as the main reaction mechanism for the C₃ products formation, which is also the most commonly used mechanism in previous reports. To screen the possible metals for doping in Cu, this work substituted one surface copper atom with Ag, Au, Pd, Rh, and Ru, some of which is shown in FIGS. 5-6. Based on the above surfaces, this work calculated the barriers of C₁-C₁ and C₁-C₂ coupling, and the results are shown in Supplementary Table 1.

To support the geometries proposed, this work calculated the CO adsorption energies on all the possible 16 sites of Ag-doped Cu surface including all of fcc hollow, hcp hollow, bridge, or top sites of Cu-a, Cu-b, and Ag (Cu-a and Cu-b determined by their coordination environment), as shown in Supplementary Table 2. By comparing all the CO adsorption energies on different sites, the configuration with CO adsorbing on hcp site of Cu-b, Cu-b, and Cu-a is the most stable and thus were used for this calculation. Due to the similarity of adsorption schemes and structures of CO, OCCO, and OCCOCO, this work assumed the strongest adsorption sites of CO are also the adsorption sites for OCCO and OCCOCO.

Additionally, this work also tested other functionals including Bayesian error estimation functional (BEEF), Perdew-Burke-Ernzerhof revised for solids (PBEsol), and Perdew-Wang 91 (PW91) for the activation energies of C₁-C₁ and C₁-C₂ coupling on Cu and Ag-doped Cu (Supplementary Table 3). Each functional predicts an enhancement of C—C coupling of Ag-doped Cu compared to pure Cu.

The effect of strain is evaluated by using the structures of all transition states and intermediate states on Ag-doped Cu surface and Cu with strain surface. The surface atoms are then fixed, and all the transition states and intermediate states are optimized based on the methods above.

Using the energetic span model, the rate of C₂ formation should be r_C2˜e^−EC1-C1/RT, and C₃ formation rate should be r_C3˜e^{−( ΔHC1-C1+EC1-C2)/RT}, where E_C1-C1 and E_C1-C2 are the barrier for C₁-C₁ and C₁-C₂ coupling, and ΔH_C1-C1 is the enthalpy change for C₁-C₁ coupling. As ΔH_C1-C1+E_C1-C2>E_C1-C1, C₃ formation rate should be slower than C₂ formation rate according to the calculation, in agreement with the experiment results.

Chemicals and Materials

Commercial Cu nanopowder (99%) and silver nitrate (AgNO₃, 99%) were purchased from Sigma-Aldrich. Potassium hydroxide (KOH) and methanol were purchased from Caledon Laboratory Chemicals. Gas diffusion layer (GDL, Freudenberg H14C9), anion exchange membrane (Fumasep FAB-PK-130) were received from Fuel Cell Store. Ni foam (1.6 mm thickness) was purchased from MTI Corporation. All chemicals were used as received. All aqueous solutions were prepared using deionized water with a resistivity of 18.2 MΩcm⁻¹.

Calculation for Equilibrium Potential

Equilibrium potentials for the half reactions of CO to n-propanol and CO₂ to n-propanol were calculated based on the values of the standard molar Gibbs energy of formation at 298.15 K (ref. 35). This work assumed that gases are at 1 atm and liquids are in the pure form.

3CO(g)+12e⁻+12H⁺→C₃H₇OH(l)+2H₂O(l) ΔG°=−231.5 kJ mol⁻¹

3CO₂(g)+18e⁻+18H⁺→C₃H₇OH(l)+5H₂O(l) ΔG°=−171.2 kJ mol⁻¹

Based on

$\begin{array}{cc}{E}^o=\frac{-\Delta G°}{\mathrm{nF}},& \left(\mathrm{equation}1\right)\end{array}$

one can get E_n-propanol-1^O=0.20 V versus RHE and E_n-propanol-2^O=0.10 V versus RHE for the reactions of CO to n-propanol and CO₂ to n-propanol, respectively. Herein, n is the number of electrons transferred and F is the Faraday constant.

Calculation for Cathodic Energy Conversion Efficiency

The OER in anode side is one of main contributors to the energy lost, but here this work excluded the effect of the OER and analyzed the cathode performance using cathodic energy conversion efficiency (EE_{cathodic half-cell}), where the overpotential of oxygen evolution is assumed to be 0.

The n-propanol EE_{cathodic half-cell} can be calculated as follows:

$\begin{array}{cc}{\mathrm{EE}}_{\mathrm{cathodic}\mathrm{half}-\mathrm{cell}}=\frac{\left(1.23+\left(-E_{n-\mathrm{propanol}}\right)\right)\times {\mathrm{FE}}_{n-\mathrm{propanol}}}{\left(1.23+\left(-E\right)\right)},& \left(\mathrm{equation}2\right)\end{array}$

where E is applied potential versus RHE, FE_n-propanol is the measured Faradaic efficiency of n-propanol in percentage, and E_n-propanol=0.20 V_RHE for CORR or E_n-propanol=0.10 V_RHE for CO₂RR. As shown in the equation above, n-propanol EE_{cathodic half-cell} is governed by both FE and overpotential, which are the two important factors in CORR.

For example, at −0.416 V_RHE after iR compensation, Ag-doped Cu GDE delivered a n-propanol FE of 33% in CORR. Then the n-propanol EE_{cathodic half-cell} for n-propanol can be calculated as follows:

${\mathrm{EE}}_{\mathrm{cathodic}\mathrm{half}-\mathrm{cell}}=\frac{\left(1.23+\left(-0.2\right)\right)\times 33\%}{(1.23+\left(-\left(-0.416\right)\right)}=20.7\%$

Supplementary References. These references are also incorporated herein by reference.

1. Kozuch, S. & Shaik, S. How to conceptualize catalytic cycles? The energetic span model Acc. Chem. Res. 44, 101-110 (2011).

2. Nørskov, J. K. et al. Origin of the overpotential for oxygen reduction at a fuel-cell cathode. J. Phys. Chem. B 108, 17886-17892 (2004).

3. Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169-11186 (1996).

4. Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comp. Mater. Sci. 6, 15-50 (1996).

5. Kresse, G. & Hafner, J. Ab initio molecular-dynamics simulation of the liquid-metal-amorphous-semiconductor transition in germanium. Phys. Rev. B 49, 14251-14269 (1994).

6. Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B 47, 558-561 (1993).

7. Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865-3868 (1996).

8. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B, 59, 1758-1775 (1999).

9. Blöchl, P. E. Projector augmented-wave method. Phys. Rev. B, 50, 17953-17979 (1994).

10. Grimme, S., Antony, J., Ehrlich, S. & Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 132, 154104 (2010).

11. Michaelides, A. et al. Identification of general linear relationships between activation energies and enthalpy changes for dissociation reactions at surfaces. J. Am. Chem. Soc. 125, 3704-3705 (2003).

12. Liu, Z.-P. & Hu, P. General rules for predicting where a catalytic reaction should occur on metal surfaces: a density functional theory study of C—H and C—O bond breaking/making on flat, stepped, and kinked metal surfaces. J. Am. Chem. Soc. 125, 1958-1967 (2003).

13. Alavi, A., Hu, P., Deutsch, T., Silvestrelli, P. L. & Hutter, J. CO oxidation on Pt(111): an ab initio density functional theory study. Phys. Rev. Lett. 80, 3650-3653 (1998).

14. Wang, Z., Liu, X.; Rooney, D. W. & Hu, P. Elucidating the mechanism and active site of the cyclohexanol dehydrogenation on copper-based catalysts: A density functional theory study. Surf. Sci., 640, 181-189 (2015).

15. Montoya, J. H., Shi, C., Chan, K. & Nørskov, J. K. Theoretical insights into a CO dimerization mechanism in CO₂ electroreduction. J. Phys. Chem. Lett. 6, 2032-2037 (2015).

16. Li, C. W., Ciston, J. & Kanan, M. W. Electroreduction of carbon monoxide to liquid fuel on oxide-derived nanocrystalline copper. Nature 508, 504-507 (2014).

17. Handoko, A. D. et al. Mechanistic insights into the selective electroreduction of carbon dioxide to ethylene on Cu2O-derived copper catalysts. J. Phys. Chem. C 120, 20058-20067 (2016).

18. Ren, D. et al. Mechanistic insights into the enhanced activity and stability of agglomerated Cu nanocrystals for the electrochemical reduction of carbon dioxide to n-propanol. J. Phys. Chem. Lett. 7, 20-24 (2016).

19. Kim, D., Kley, C. S., Li, Y. & Yang, P. Copper nanoparticle ensembles for selective electroreduction of CO₂ to C₂-C₃ products. Proc. Natl. Acad. Sci. U. S. A. 114, 10560-10565 (2017).

20. Rahaman, M., Dutta, A., Zanetti, A. & Broekmann, P. Electrochemical reduction of CO₂ into multicarbon alcohols on activated Cu mesh catalysts: an identical location (IL) study. ACS Catal. 7, 7946-7956 (2017).

21. Jiang, K. et al. Metal ion cycling of Cu foil for selective C—C coupling in electrochemical CO₂ reduction. Nat. Catal. 1, 111-119 (2018).

22. Zhuang, T.-T. et al. Steering post-C—C coupling selectivity enables high efficiency electroreduction of carbon dioxide to multi-carbon alcohols. Nat. Catal. 1, 421-428 (2018).

23. Verdaguer-Casadevall, A. et al. Probing the active surface sites for CO reduction on oxide-derived copper electrocatalysts. J. Am. Chem. Soc. 137, 9808-9811 (2015).

24. Feng, X., Jiang, K., Fan, S. & Kanan, M. W. A direct grain-boundary-activity correlation for CO electroreduction on Cu nanoparticles. ACS Cent. Sci. 2, 169-174 (2016).

25. Raciti, D. et al. Low-overpotential electroreduction of carbon monoxide using copper nanowires. ACS Catal. 7, 4467-4472 (2017).

26. Jouny, M., Luc, W. & Jiao, F. High-rate electroreduction of carbon monoxide to multi-carbon products. Nat. Catal. 1, 748-755 (2018).

27. Li, J. et al. Copper adparticle enabled selective electrosynthesis of n-propanol. Nat. Commun. 9, 4614 (2018).

28. Zhuang, T.-T. et al. Copper nanocavities confine intermediates for efficient electrosynthesis of C3 alcohol fuels from carbon monoxide. Nat. Catal. 1, 946-951 (2018).

29. Xiao, H., Cheng, T. & Goddard, W. A. Atomistic mechanisms underlying selectivities in C₁ and C₂ products from electrochemical reduction of CO on Cu (111). J. Am. Chem. Soc. 139, 130-136 (2016).

30. Kuhl, K. P., Cave, E. R., Abram, D. N. & Jaramillo, T. F. New insights into the electrochemical reduction of carbon dioxide on metallic copper surfaces. Energy Environ. Sci. 5, 7050-7059 (2012).

31. Wellendorff, J. et al. Density functionals for surface science: Exchange-correlation model development with Bayesian error estimation. Phys. Rev. B 85, 235149 (2012).

32. Perdew, J. P. et al. Restoring the density-gradient expansion for exchange in solids and surfaces. Phys. Rev. Lett. 100, 136406 (2008).

33. Perdew, J. P. & Wang, Y. Accurate and simple analytic representation of the electron-gas correlation energy. Phys. Rev. B 45, 13244 (1992)

34. Kozuch, S. & Shaik, S. How to conceptualize catalytic cycles? The energetic span model. Acc. Chem. Res. 44, 101-110 (2011).

35. Rumble, J. R. CRC Handbook of Chemistry and Physics 99th edn (CRC Press, Cleveland, 2018).

36. Dinh, C.-T. et al. CO₂ electroreduction to ethylene via hydroxide-mediated copper catalysis at an abrupt interface. Science 360, 783-787 (2018).

In addition, it is noted that any of the particle values disclosed herein can be considered as being ±10% for disclosure purposes. Thus, for example, if a concentration value of 1 g/L is mentioned herein, it should be considered that the range 0.9 to 1.1 g/L is disclosed. It is also noted that one or more features (e.g., values, ranges, pieces of equipment or features thereof, operating conditions, sizes, etc.) disclosed herein can be combined with any other combination of features. For example, if a size of 100 nm is disclosed for a certain nanoparticle herein, it should be noted that the multimetallic (e.g., bimetallic or trimetallic) nanoparticle catalyst material disclosed herein can be, in an optional embodiment, of this size (i.e., 100 nm in addition to within the range of 90 nm to 110 nm as per the ±10% disclosure). Various other combinations of features are also possible and should be considered as being disclosed herein.

Claims

1. An electrocatalyst for electroreduction of a carbon-containing gas to produce C3 products, the electrocatalyst comprising a multi-metallic material comprising a primary metal and a metal dopant selected and distributed to provide asymmetric active sites that include neighbouring atoms of the primary metal having distinct electronic structures to promote C2-C1 coupling.

2. The electrocatalyst of claim 1, wherein the primary metal is Cu.

3. The electrocatalyst of claim 2, wherein the metal dopant is Ag and the multi-metallic material is a bimetallic material.

4. The electrocatalyst of claim 1, wherein the metal dopant comprises Ag and a second dopant metal and the multi-metallic material is a trimetallic material.

5-7. (canceled)

8. The electrocatalyst of claim 1, wherein the C3 product is n-propanol.

9. The electrocatalyst of claim 1, wherein the multi-metallic material comprises the primary metal doped with the metal dopant using galvanic replacement.

10. The electrocatalyst of claim 1, wherein the metal dopant is present in the primary metal in a doping concentration of 2 wt % to 9 wt %, measured with XPS.

11. (canceled)

12. The electrocatalyst of claim 1, wherein the metal dopant is present in the primary metal in a doping concentration of 3wt % to 5 wt %, measured with XPS.

13-14 (canceled)

15. The electrocatalyst of claim 1, wherein the electrocatalyst is provided in the form of bimetallic or trimetallic nanoparticles.

16. The electrocatalyst of claim 15, wherein the bimetallic or trimetallic nanoparticles have an average size between about 20 nm and about 200 nm, measured based on SEM or TM imaging. Claims 17-18 (canceled)

19. The electrocatalyst of claim 16, wherein the bimetallic or trimetallic nanoparticles are generally spheroid in shape, determined from SEM or TM imaging.

20. The electrocatalyst of claim 1, wherein the electrocatalyst is formed as a deposited catalyst layer on a first side of gas diffusion membrane, wherein the deposited catalyst layer is configured to be in direct contact with an electrolyte and wherein a second opposed side of the gas diffusion membrane is configured to be in direct contact with the carbon-containing gas.

21. An electrocatalyst for electroreduction of a carbon-containing gas to produce a C3 product, the electrocatalyst comprising a bimetallic or trimetallic material comprising a copper (Cu) and at least one metal dopant in a doping concentration of 2 wt % to 9 wt %, measured with XPS.

22. (canceled)

23. The electrocatalyst of claim 21, wherein the metal dopant comprises a primary dopant and a secondary dopant.

24. The electrocatalyst of claim 23, wherein the primary dopant is Ag.

25. The electrocatalyst of claim 24, wherein the secondary dopant is Ru.

26-37. (canceled)

38. A method of fabricating an electrocatalyst composed of a primary metal and at least one metal dopant for electroreduction of a carbon containing gas into C3 products, such as n-propanol, comprising galvanic replacement, wherein the electrocatalyst is as defined in claim 1.

39. The method of claim 38, comprising:

providing a layer of Cu particles on a substrate to provide a coated substrate;

immersing the coated substrate in an Ag containing aqueous solution to induce doping and form an Ag-doped multimetallic catalyst material supported by the substrate; and

removing the coated substrate from the solution, the coated substrate comprising a layer of the Ag-doped multimetallic catalyst material.

40-44. (canceled)

45. A process for electrochemical production of a C3 multi-carbon compound from a carbon-containing gas, comprising:

contacting the carbon-containing gas and an electrolyte with an electrode comprising the electrocatalyst as defined in claim 1, such that the carbon-containing gas contacts the electrocatalyst;

applying a voltage to provide a current density to cause the carbon-containing gas contacting the electrocatalyst to be electrochemically converted into the C3 multi-carbon compound; and

recovering the C3 multi-carbon compound.

46-47. (canceled)

48. The process of claim 45, wherein the C3 multi-carbon compound is n-propanol.

49. (canceled)

50. The process of claim 45, wherein the electrolyte comprises KOH.

51. The process of claim 45, wherein the carbon-containing gas comprises CO.

52. The process of claim 45, wherein the carbon-containing gas comprises CO2.

53. (canceled)

54. A system for CO and/or CO2 electroreduction to produce a multi-carbon compound, comprising:

an electrolytic cell configured to receive a liquid electrolyte and CO and/or CO2 gas;

an anode;

a cathode comprising an electrocatalyst as defined in claim 1; and

a voltage source to provide a current density to cause the CO and/or CO2 gas contacting the electrocatalyst to be electrochemically converted into the multi-carbon compound.

55. (canceled)