Enhanced CO2 Electroreduction to Multi‐Carbon Products on Copper via Plasma Fluorination

Abstract The electroreduction of carbon dioxide (CO2) to multi‐carbon (C2+) compounds offers a viable approach for the up‐conversion of greenhouse gases into valuable fuels and feedstocks. Nevertheless, current industrial applications face limitations due to unsatisfactory conversion efficiency and high overpotential. Herein, a facile and scalable plasma fluorination method is reported. Concurrently, self‐evolution during CO2 electroreduction is employed to control the active sites of Cu catalysts. The copper catalyst modified with fluorine exhibits an impressive C2+ Faradaic efficiency (FE) of 81.8% at a low potential of −0.56 V (vs a reversible hydrogen electrode) in an alkaline flow cell. The presence of modified fluorine leads to the exposure and stabilization of high‐activity Cu+ species, enhancing the adsorption of *CO intermediates and the generation of *CHO, facilitating the subsequent dimerization. This results in a notably improved conversion efficiency of 13.1% and a significant reduction in the overpotential (≈100 mV) for the C2+ products. Furthermore, a superior C2+ FE of 81.6% at 250 mA cm−2, coupled with an energy efficiency of 31.0%, can be achieved in a two‐electrode membrane electrode assembly electrolyzer utilizing the fluorine‐modified copper catalyst. The strategy provides novel insights into the controllable electronic modification and surface reconstruction of electrocatalysts with practical potential.

The polytetrafluoroethylene (PTFE) with a pore size of 1 μm was used as the substrate and gas diffusion layer.Approximately 500 nm Cu were evaporated on the substrate in ~10 −4 Pa at 0.4 Å/sec.

Fabrication of F-Cu catalyst
The Cu catalyst was treated by CF 4 plasma (Inductively Coupled Plasma etching system from Leuven, ICP) for 1 minute with power of 200 W and 50 W RF bias at 5 mTorr and 20 °C (F x -Cu).Then, the F x -Cu pre-catalyst was reduced in 1 M KOH solution at −0.8 V versus RHE for 10 min to fabricate the F-Cu catalyst.The F-Cu catalyst under different powers can also be prepared, such as F-Cu-100 W and F-Cu-150 W by changing the plasma power.

Characterization
Phase components of the as-prepared catalysts were tested by X-ray diffractometer (XRD, Germany Bruker, D8 Advance) with Cu Kα radiation (λ = 0.15418 nm).Morphologies of the catalysts were confirmed by a field emission scanning electron microscope (FESEM, Hitachi High-Technologies Corporation, Regulus 8230) and a transmission electron microscope (TEM, Thermo Fisher Scientific, Talos F200X G2) equipped with energy-dispersive X-ray detectors (Oxford, Ultim EXTREME).X-ray photoelectron spectroscopic (XPS, Thermo Fisher Scientific, ESCALAB Xi+) was employed to detect the surface chemical state of samples.Electron paramagnetic resonance (EPR) measurements were carried out on a CIQTEK EPR200M spectrometer and microwave frequency = 9.40 GHz at room temperature.CO 2 sorption isotherms were measured using Micromeritics 3FLEX at 25 °C.TPD measurements were performed using Microtrac BELCat II.

Electrochemical measurements
The CO 2 RR in the flow cell (Gaoss Union, 1 cm 2 ) separated by an anion-exchange membrane (FAA-3-PK-130, Fumasep) was controlled by a CHI 660e electrochemical workstation.The as-made catalysts were directly employed as the working electrode with Ag/AgCl electrode (reference electrode, 3 M KCl) and nickel foam (counter electrode).All the potentials were converted to values with reference to the RHE using: E (RHE) = E (Ag/AgCl) + 0.21 V + 0.059 V × pH + iR, with 85% compensation.50 mL 1 M KOH was used as catholyte and anolyte.The electrolytes in the cathode and anode were circulated by two pumps at the rate of 5 ml min −1 and 90 ml min −1 , respectively.Meanwhile, CO 2 gas was continuously supplied to the gas chamber of the cathode at the rate of 30 ml min −1 .The performance of the cathodes was evaluated by performing constant-current electrolysis.Gas and liquid products were analyzed using a gas chromatograph (GC, Shimadzu, GC-2014) equipped with thermal conductivity and flame ionization detectors and a nuclear magnetic resonance (NMR) spectrometer (Bruker BioSpin, AVANCE NEO) by taking dimethylsulfoxide (DMSO) as an internal standard, respectively.
For the electrochemical CO 2 RR test in an MEA electrolyzer, a commercial MEA electrolyzer (Gaoss Union, 1 cm 2 ) was used.The F-Cu, an anion-exchange membrane (Sustainion® X37-50) and nickel foam were compressed to form MEA. 1.0 M KOH solution was served as the anolyte with a flow rate of 90 ml min −1 .The flow rate of cathodic CO 2 gas was kept at 30 sccm and flowed through a homemade humidifier (deionized water, room temperature) before the MEA.The electrolysis of the MEA electrolyzer was performed on a CHI 1140c electrochemical workstation.
The Faradaic efficiencies (FEs) were calculated on the basic of the following equation: where Qx and Q total were the charge passed into product x and passed charge (C) during CO 2 RR, n x represents the electron transfer number of product x, Nx was the product amount (mol) of x measured by GC or NMR and F was the Faraday constant (96485 C mol −1 ).
The energy efficiencies (EEs) were calculated on the basic of the following equation: where E applied represents the potential applied during the CO 2 RR, E x 0 is the thermodynamic potential (vs.RHE) for the product x, such as 0.08 V for ethylene formation, 0.09 V for ethanol formation, 0.1 V for propanol formation and 0.11 V for acetic acid formation. [1]  In-situ Raman measurements Raman spectroscopy was carried out in a custom-built flow cell using a Raman spectrometer (WITec Alpha 300R).A 785-nm laser was used with a laser power of 3 mW, and signals were recorded using a 5-s integration and by averaging two scans.A 10× objective lens was used for focusing and collecting the incident and scattered laser light.The electrochemical CO 2 reduction was performed at different currents from 0 to −15 mA cm −2 in 1 M KHCO 3 .

Computational Methods
All calculations are performed in the framework of the density functional theory with the projector augmented plane-wave method, as implemented in the Vienna ab initio simulation package. [2]The generalized gradient approximation proposed by Perdew, Burke, and Ernzerhof is selected for the exchange-correlation potential. [3]The long-range van der Waals interaction is described by the DFT-D3 approach. [4]The cut-off energy for the plane wave is set to 450 eV.The energy criterion is set to 10 −5 eV in the iterative solution of the Kohn-Sham equation.A large supercell (~10 Å× 10Å×21 Å) was adopted to simulate the Cu(111) and F-Cu(111) slab.An F atom was anchored on the surface through F-Cu bonds at the hollow site.A vacuum layer of 15 Å is added perpendicular to the sheet to avoid artificial interaction between periodic images.All the structures are relaxed until the residual forces on the atoms have declined to less than 0.03 eV/Å.During geometry relaxation, the bottom two layers were relaxed.For the reaction free energy calculations, a computational hydrogen electrode model was employed. [5]Gibbs free energies (G) were estimated as G = E DFT + G ZPE + ∆TS.Here, E DFT is the electronic energy computed from DFT calculations.

Figure S7 .
Figure S7.EPR spectra of Cu and F-Cu catalysts.

Figure S10 .Figure S11 .
Figure S10.F 1s XPS spectra for (a) F x -Cu-100 and (b) F x -Cu-150 with respect to different Ar + beam etching depths.The etching depths are estimated values based on theoretical parameters.These numbers of 100, 150 and 200 represent the different powers in the plasma process.These samples were directly fluorinated by CF 4 plasma using copper foil as the precursor.

Figure S12 .
Figure S12.F 1s XPS spectra for (a) F-Cu-100, (b) F-Cu-150, and (c) F-Cu-200 with respect to different Ar + beam etching depths.These numbers of 100, 150 and 200 represent the different powers in the plasma process.These samples were directly fluorinated by CF 4 plasma using copper foil as the precursor.

Figure S15 .
Figure S15.LSV curves in Ar and CO 2 environments at a scan rate of 50 mV s −1 over F-Cu catalyst and Cu catalyst in 1.0 M KOH.

Figure S16 .
Figure S16.Representative nuclear magnetic resonance (NMR) spectra of the liquid products for F-Cu catalyst (red) and Cu catalyst (blue) at a current density of 100 mA cm −2 .

Figure S17 .
Figure S17.Comparison of electrochemical CO 2 RR performance on F-Cu catalyst with reported Cu-based catalysts in the flow cell.

Figure S18 .
Figure S18.(a) Stability test of the F-Cu catalyst at a current density of 200 mA cm −2 and 150 mA cm −2 .(b) LSV curves of the F-Cu catalyst before and after the 10 h stability test as well as the LSV curve after the consequent electrolyte refresh.

Figure S19 .
Figure S19.(a) SEM image and HRTEM image of F-Cu catalyst after 12h electrochemical CO 2 RR.

Figure S20 .
Figure S20.(a) F 1s XPS spectra and (b) Cu LMM Auger spectra of F-Cu catalyst before and after 12 h electrochemical CO 2 RR.

Figure S21 .
Figure S21.(a) The cyclic voltammetry and (b) double-layer capacitance curves of the Cu catalyst.(c) The cyclic voltammetry and (d) double layer capacitance curve of the F-Cu catalyst.Generally, ECSA = R f × S, where R f was calculated from the ratio of double-layer capacitance (C dl ) for the working electrode and the corresponding smooth polycrystalline Cu electrode (29 μF cm −2 ) and S stands for the geometric area of the electrode (in this work, S = 1 cm 2 ).Therefore, the ECSA are 16.24 and 19.1 cm 2 for the Cu and F-Cu catalysts, respectively.

Figure S22 .
Figure S22.(a) The cyclic voltammetry and (b) double layer capacitance curves of the F-Cu-100 W catalyst.(c) The cyclic voltammetry and (d) double layer capacitance curve of the F-Cu-150 W catalyst.

Figure S23 .Figure S24 .
Figure S23.Nyquist plots of Cu catalyst and F-Cu catalyst in 1.0 M KOH at −0.2 V vs. RHE.

Figure S25 .
Figure S25.CO 2 sorption isotherms at 273 K of Cu and F-Cu catalysts.

Figure S26 .
Figure S26.The DFT models of a Cu (111) surface with fluorine adsorption at

Figure S27 .
Figure S27.The DFT models of *CO adsorption on different F-Cu surfaces with various concentrations of fluorine.

Figure S31 .
Figure S31.LSV curve at a scan rate of 50 mV s −1 over F-Cu catalyst in the MEA system.

Table S1 .
Comparison of Electrocatalytic performances for CO 2 to C 2+ products on F-Cu catalyst in a flow cell with reported values on Cu-based catalysts.

Table S2 .
Comparison of Electrocatalytic performances for CO 2 to C 2+ products on F-Cu catalyst with reported values on Cu-based catalysts modified by nonmetallic heteroatoms.