Materials preparationPristine Ag foil electrode: The Ag foil (>99.99%, Wuhu Yanjiao Co., LTD) was pretreated by 0.01 M hydrochloric acid, acetone, ethanol, deionized water (DI, 18.2 MΩ) obtained from an ultra-pure purification system (Master-S15Q, Hitech Instruments Co. Ltd., Shanghai, China), and then dried with a stream of nitrogen. The treated Ag foil was cut into a square with a geometric area of 1 cm2, which was used as the pristine Ag foil electrode.AC-Ag: The Ag foil electrode was activated in 0.5 M KCl aqueous solution by cyclic voltammetry within the potential window of 0.05 V to 1.05 V (vs. RHE). The scan rate is 10 mV s−1. After five cycles, a pale-yellow electrode was obtained, which is denoted as AC-Ag. The in situ-activated AC-Ag was used as a working electrode for Br-EO reduction without any treatment.AC-Ag/C catalyst: Ag/C catalyst was prepared by wet chemical method. Silver nitrate (AgNO3, 0.17 g) was added into 20 mL oleamine (70%) containing 0.1 mL oleic acid. The solution was heated to 180 oC under the protection of N2, keeping this temperature for 2 h. Then the reaction was stopped and natural cooling. The precipitate was separated by centrifugal and washed with acetone five times. The obtained sample was dispersed in 30 mL hexane. To achieve an Ag/C catalyst, commercial carbon black (0.25 g) was mixed with the Ag nanoparticle dispersions by sonication. Finally, Ag/C can be collected after centrifugation and vacuum drying. The obtained Ag/C undergoes the same electrochemical activation step as AC-Ag preparation (cyclic voltammetry scanning from 0.05 to 1.05 V (vs RHE) with five cycles, at 10 mV s−1).Cu/Cu2O catalyst: 0.1 M copper chloride (20 mL) was added dropwise to sodium hydroxide solution (0.2 M, 30 mL) under the continuous N2 bubbling. Then 200 μL ethanol and 25 mL sodium borohydride solution (0.05 M) were poured quickly. After vigorous stirring for 0.5-h, Cu/Cu2O was obtained by centrifugal and rinsed with DI water, then following a vacuum drying step at 80 oC.Physical characterizationThe phase detections of the samples were performed on a D/max 2550 VB X-ray diffractometer using Cu Kα radiation (λ = 0.154 nm) and the scan speed was 10o min−1. X-ray photoelectron spectra (XPS) were measured using an ECSALAB250Xi spectrometer with an Al Kα X-ray (1486.6 eV) radiation for excitation, and the binding energy was corrected by C 1 s value of 284.6 eV. Transmission electron microscopy (TEM) with a spherical aberration corrector (HRTEM, Titan G2 60-300) was used for morphology measurements. The surface morphology of Ag foil and AC-Ag was observed by scanning electron microscopy (JEOL-6701F) at an acceleration voltage of 5 kV. The ion component in the electrolyte was tested by ion chromatography (Thermo Scientific Aquion). In situ electrochemical FTIR spectroscopic was performed using a Fourier transform infrared spectrometer (Nicolet iS50), coupling with an in-situ electrochemical reaction cell (SPEC-I, Yuanfang Co. Ltd., Shanghai, China). The catalyst ink, obtained by dispersing Ag/C in ethanol solution (950 μL) containing 50 μL Nafion binder (5 wt% aqueous solution), was dripped onto a hemicylindrical silicon prism covered with a layer of gold membrane. A platinum wire and Ag/AgCl electrode were used as counter and reference electrodes. The electrolyte was 0.5 M KCl with or without Br-EO. During the test, the electrolyte was constantly purged with CO2. The background spectrum (reflectance R0) was recorded at open circuit voltage. All spectra were reported as the relative change in reflectivity, ΔR/R0 = (R-R0)/R0. The R and R0 are single-beam spectra collected at the applied bias and the reference potential. Operando Raman measurements were performed using a Raman microscope (DXR2, thermo scientific) with a 532 nm laser. The measurements were carried out in a homemade three-electrode electrochemical cell. Ag/AgCl (saturated KCl) and Pt wire were used as the reference and counter electrodes, respectively. The working electrode was a glassy carbon electrode, which was connected to an electrochemical workstation (760E) and laid flat on the microscope stage. In the process of each experiment, the electrolyte was continuously circulated by a peristaltic pump.Electrochemical measurementAll electrochemical tests were performed by a Biologic SP-300 workstation. A typical three-electrode system in a custom-made gastight H-type two-compartment cell (30 mL, Aida Technology Development Co., Ltd, Tianjin, China) was accepted for 2-bromoethanol (Br-EO) reduction. Pristine Ag foil, AC-Ag, or carbon paper deposited Ag/C catalyst was used as a working electrode, and their geometric areas are all 1.0 cm2. The mass loading of AC-Ag/C is 1 mg cm−2. The reference and counter electrodes were Ag/AgCl (leak-free, Aida Technology Development Co., Ltd, Tianjin, China) and Pt mesh electrodes (1 × 1 cm2). The electrolyte was 0.5 M KCl (pH = 5.49 ± 0.07). Br-EO was added to the catholyte with concentrations of 50 mM, 25 M, 10 mM, 5 mM, and 2 mM. A proton exchange membrane (Nafion 117, 2.0 × 2.0 cm, 118.0 μm) was placed between the cathode and anode chamber as a separator. The catholyte was purged with CO2 or N2 gas (99.999%) for at least 30 min, and the gas was constantly bubbled through the catholyte during electrolysis to flow the reductive product into gas chromatography (GC), at a flow rate of 30 sccm which was controlled by a mass flow controller (HORIBA METRON). Linear sweep voltammetry and potentiostatic electrolysis were conducted with the IR compensation of 85% in situ. To avoid the electrooxidation of metallic Ag, the onset potential for the linear sweep voltammetry was set as −0.25 V, which is far lower than the oxidation potential of Ag. The measurement in 0.5 KBr (pH = 5.47 ± 0.1) was performed once. The stability of the catalyst was evaluated by injecting Br-EO into the electrolyte after the ethylene selectivity dropped. As for the concentration dependence of Br-EO, KCl, and H2O on the activity of ethylene formation, potentiostatic electrolysis was performed at −0.28, −0.38, and −0.48 V. To quench the possible generated *H, 50 mM n-butyl alcohol was added to 0.5 M KCl containing 50 mM Br-EO and the selectivity and current density was monitored at −0.38 and −0.48 V. Herein, all potentials were converted to RHE reference scale using the following equation:$$E({{{\rm{vs\; RHE}}}})=E({{{\rm{vs\; Ag}}}}/{{{\rm{AgCl}}}})+0.197{{{\rm{V}}}}+0.0591\times {{{\rm{pH}}}}$$
(1)
The electrochemical active surface area (ECSA) was analyzed by Pd underpotential deposition. The electrolyte consists of a KNO3 (0.1 M)/HNO3 (0.01 M) aqueous solution containing 5 mM Pb(ClO4)2. The theoretical deposition potential for Pb2+ is calculated to be −0.14 V (vs RHE) based on the Nernst equation. A typical three-electrode system was employed with the reference electrode and counter electrode of Ag/AgCl and Pt mesh (1 × 1 cm2), respectively.The activation energy of Br-EO reduction was acquired by linear fitting of the natural logarithm of ethylene partial current density versus the inverse temperatures based on the Arrhenius equation.$${{{\mathrm{ln}}}}\left({{\mbox{i}}}\right)=\frac{-{{{{\rm{E}}}}}{\alpha }}{{RT}}+{{\mathrm{ln}}}\left({\mbox{A}}\right)$$
(2)
Where i is the partial current density, Eα is the activation energy, R is the gas constant, T is the reaction temperature, and A is a pre-exponential factor.Electrochemical Impedance spectroscopy (EIS) was collected by a three-electrode cell. The measurements were conducted at constant potentials in the frequency range from 50 MHz to 100 KHz with an AC amplitude of 10 mV. Then the relaxation time constant (τ0) was acquired by complex capacitance (C’) analysis.$$C^{\prime} ({{{\rm{\omega }}}})=\frac{-Z^{\prime\prime} ({{{\rm{\omega }}}})}{{{{\omega }}{\mbox{|}}Z\left({{{\rm{\omega }}}}\right){\mbox{|}}}^{2}}$$
(3)
$$C^{\prime\prime} \left({{{\rm{\omega }}}}\right)=\frac{Z^{\prime} ({{{\rm{\omega }}}})}{{{{\omega }}{\mbox{|}}Z\left({{{\rm{\omega }}}}\right){\mbox{|}}}^{2}}$$
(4)
$${{\mbox{Z}}}\left({{{\rm{\omega }}}}\right)=\frac{1}{j{{\omega }}C\left({{{\rm{\omega }}}}\right)^{\prime} }$$
(5)
Where Z(ω) is impedance and ω is the penetration depth. C′(ω) and C′′(ω) are the real part and imaginary part of the capacitance. According to the peak frequency (f0), the τ0 can be calculated using the equation of τ0 = (2πf0)−1.The transient kinetics measurement was collected by pulse amperometry. The adsorption of Br-EO was conducted at −0.08 V (vs RHE) for 100 s, followed by the application of a pulse potential of −0.28 V (vs RHE) for 10 s. The time duration of 0–2 s was analyzed for the transient kinetics.The electrical work was calculated via the following equation.$${W}_{{{\rm{Br}}}-{{\rm{EO}}}}=U\times \frac{{10}^{6}}{{M}_{{{\rm{ethylene}}}}}\times \frac{F\times N}{{{{\rm{FE}}}}_{{{\rm{Br}}}-{{\rm{EO}}}}}$$
(6)
$${W}_{{{\rm{ethylene}}}}=U\times \frac{{10}^{6}\times F\times N}{{M}_{{{\rm{ethylene}}}}\times {{\mbox{FE}}}_{{{\rm{ethylene}}}}}$$
(7)
Where U is cell potential which was collected by two-electrode measurement using constant current mode, Methylene is the molecular weight of ethylene. F is the Faradaic constant. N is the number of electon transferred.The CO2 electroreduction activity was measured by a cation-exchange membrane-separated flow cell. The catholyte is 1 M KOH (pH = 13.35 ± 0.09) and the anolyte is 1 M KBr (pH = 5.07 ± 0.03). Cu/Cu2O was first dispersed in ethanol and ultrasonic dispersion to give an ink, which was further spray coated onto a gas diffusion electrode (Sigracet 35 BC) with a geometric surface area of 2 × 3 cm2. The mass loading of the catalyst was 1 mg cm−2. The Pt plate (2 × 3 cm2) and Ag/AgCl electrode acted as counter electrode and reference electrode, respectively. During the electrochemical testing, CO2 flowed over the cathode GDL with a flow rate of 50 sccm, and the electrolyte was circulated at a flow rate of 1 mL min−1. The outlet gas of the cathode flowed through the anolyte jar to convert the ethylene in the gas mixture to Br-EO. After CO2RR testing, the alkaline catholyte was gradually added into the anolyte to remove the residue Br2 and neutralize the formed acid, until a neutral solution formed, and then the solution was used for further reduction by the Ag/C catalyst in an H-type cell. The combined measurement for indirect CO2-to-ethylene conversion was performed once as the proof of concept.Product analysisThe reduction products were analyzed by an on-line gas chromatograph (GC, Shimadzu, Model 2014) and 1H NMR (Bruker, 400 M Hz) spectroscopy. For gas detection, the products were brought into the GC by CO2 or N2 with the flow rate of 30 sccm which was controlled by a mass flowmeter. The GC is equipped with one TCD detector for hydrogen and CO, one flame ionization detector (FID) coupled with a methanizer for CO and CH4 detection, and one FID for C2+ chemicals. The carrier gas was Ar (99.999%). The products collected at 1000 s were sampled into the gas sampling loop of GC (1 mL). The Faradaic efficiency and partial current density were calculated as below:$${{\mbox{FE}}}_{{{\rm{x}}}}=\frac{{nF}{P}_{0}}{{RT}}\times \frac{1}{{{\alpha }}_{x}}\times {{{\rm{peak\; area}}}}\times {{{\rm{flow\; rate}}}}\times \frac{1}{I}$$
(8)
$${J}_{{{{\rm{x}}}}}={{{\rm{F}}}}{{{{\rm{E}}}}}_{{{{\rm{x}}}}}\times {j}_{{{{\rm{total}}}}}$$
(9)
Where n is the electron transfer number. F is faraday constant (96485 C mol−1) and R is gas constant (8.314 J mol−1 K−1). P0 is 1.013 bar and T0 is 273.15 K. α is conversion factors obtained based on calibration of the GC with a standard sample.The purity of ethylene specified on the mass basis was calculated by the following equation.$${P}_{{{\rm{ethylene}}}}({{{\rm{wt}}}} \% )=\frac{{{\mbox{FE}}}_{{{\rm{ethylene}}}}\times {M}_{{{\rm{ethylene}}}}}{{{\mbox{FE}}}_{{{\rm{ethylene}}}}\times {M}_{{{\rm{ethylene}}}}+{{\mbox{FE}}}_{{{\rm{H}}}2}\times {M}_{{{\rm{H}}}2}}$$
(10)
Where M is the molar molecular weight. The obtained Pethylene was corrected by the vapor in the product stream (Supplementary Note 11).For 1H NMR measurements, standards with known concentrations of products (0.25, 0.5, 1, 1.5, and 2.0 mM) were prepared in 1 mL 0.5 M KCl solution which contains 0.2 mL D2O and a known concentration of dimethyl sulfoxide (DMSO, DMSO: H2O = 1:1000 (V: V)) which acting as an internal reference. For the sample preparation, typically, after a bulk electrolysis experiment, 0.5 mL electrolyte, 0.1 mL D2O, and 0.1 mL DMSO/H2O were mixed in an NMR tube. The NMR was performed with the water suppression method.$${{\mbox{FE}}}_{{{\rm{y}}}}=\frac{{Q}_{{{\rm{y}}}}}{{Q}_{{{\rm{total}}}}}\times 100 \%=\frac{{n}_{{{\rm{y}}}}\times N\times F\times 100 \% }{{Q}_{{{\rm{total}}}}}$$
(11)
n is the amount of product. Qtotal is the total transfered charge.DFT calculationsDensity functional theory (DFT) calculations were performed with the spin-polarized generalized gradient approximation (GGA) of Perdew, Burke, and Ernzerhof (PBE) as implemented in the Vienna Ab initio Simulation Package (VASP)56,57. The van der Waals (vdW) energy correction was carried out with the DFT-D3 method58. Core electrons were handled using the projector augmented wave (PAW) method with an energy cutoff of 450 eV59. The electronic energy was considered self-consistent when the energy change was smaller than 10−5 eV. The geometries were relaxed until the energy difference was smaller 0.02 eV Å−1. During the relaxation, the Brillouin zone with a 2 × 2 × 1 Gamma-centered grid was used. A vacuum layer of 15 Å along the c-axis was used to eliminate the artificial interactions between periodic images. Spin-polarized calculations were performed for this calculation. The models based on the (111), (200), and the corresponding Cl-incorporated plane were built for free energy calculations.The reaction mechanisms for ethylene generation are used below.$${{{\rm{HO}}}}-{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}-{{{\rm{Br}}}}+\ast \to*{{{\rm{OH}}}}{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}{{{\rm{Br}}}}$$$$*{{{\rm{HO}}}}{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}{{{\rm{Br}}}}+{{{{\rm{e}}}}}^{-}\to*{{{\rm{OH}}}}{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}+{{{{\rm{Br}}}}}^{-}$$$$*{{{\rm{HO}}}}{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}+{{{{\rm{e}}}}}^{-}\to*{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2} \ast+{{{{\rm{OH}}}}}^{-}$$$$*{{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}*\to {{{{\rm{CH}}}}}_{2}{{{{\rm{CH}}}}}_{2}+\ast$$The asterisk denotes the site of the catalyst and surface adsorption.
High-purity ethylene production via indirect carbon dioxide electrochemical reduction
