Cu-Doped CeO₂ Supported by MXene Nanosheets for Enhanced Electrosynthesis of Urea from Carbon Dioxide and Nitrate

Abstract

Electrocatalytic synthesizing of urea through C-N coupling of CO₂ and NO₃⁻ under ambient conditions is a possible solution for the problem of energy consumption in commercial urea production. Herein, we report a Cu-doped CeO₂ catalyst anchored on delaminated MXene two-dimensional surface. The Cu-CeO₂/MXene catalyst achieves the co-reduction of CO₂ and NO₃⁻ to synthesize urea, obtaining a urea yield rate of 505.1 μg·h⁻¹·mg_cat.⁻¹ with a Faradic efficiency (FE) of 6.3% at −0.8 V versus reversible hydrogen electrode (vs. RHE). Theoretical calculations further demonstrate that Cu doping is capable of enhancing the activity of Cu-Ce sites and promoting C-N coupling and protonation reactions.

1. Introduction

Urea, as the first organic substance synthesized artificially, is a popular nitrogenous fertilizer because of its high content of nitrogen (about 46%) [1]. It accounts for 70% of the world’s nitrogenous fertilizer production and is one of the most lucrative industrial products of carbon dioxide derivatives. However, the current commercial urea production process relies on extreme conditions of high temperature (150~210 °C) as well as pressure (150~250 bar) [2]. It requires centralized and complex equipments and multi-cycle processes to improve the efficiency of production [3], which consumes a large amount of energy [4]. Harsh reaction conditions and huge energy consumption prevents urea production meeting the demand for sustainable development. Therefore, there is an urgent need to find an alternative technology for urea synthesis under milder conditions.

Previous research has demonstrated that electrocatalyzing C-N coupling is a practicable method that can realize the synthesis of urea under ambient conditions [5,6,7,8]. For example, Chen’s group [9] creatively introduced N₂ as nitrogen sources to synthesize urea. They accomplished the direct C-N coupling between N₂ and CO₂ via an alloy state PdCu/TiO₂ catalyst. Although copper has been validated as an effective catalyst material for electrocatalytic CO₂ reduction, the use of the high-cost material Pd still limits the practical application of this catalyst. Cao et al. [10] adulterated Cu in TiO₂ to make a Cu-TiO₂ catalyst with abundant defect sites and oxygen vacancies (V_o), obtaining good selectivity for urea electrocatalytic synthesis via the synergistic effect of Cu-Ti. This provides the idea of defect engineering.

CeO₂ is an n-type semiconductor with high electron density that is capable of stabilising metal dopants [11]. Moreover, CeO₂ is one of the richest reserves of rare earth oxides, which makes it an ideal and economical catalyst material [12]. Hence, Zhan et al. [13] used CeO₂ as a carrier and constructed CuAu single-atom alloy on it. In situ experiments and theoretical calculations showed that the Cu-Au bimetallic sites could realize the synthesis of urea by C-N coupling. The prepared Cu₁Au₈@CeO₂ catalyst had a urea yield rate of 813.6 μg·h⁻¹·mg_cat.⁻¹, achieving a superior FE of 45.2% at −0.94 V vs. RHE. Wei et al. [14] designed a V_o-rich CeO₂ catalyst and obtained a great urea yield rate of 934.6 μg·h⁻¹·mg_cat.⁻¹. After that, Wei’s team [15] constructed a catalyst loaded with copper on CeO₂ carriers, which further increased the yield rate (3173.6 μg·h⁻¹·mg_cat.⁻¹). Due to its unique two-dimensional nanostructure, MXene (transition metal carbides, nitrides, and carbonitrides) has many attractive properties such as excellent electrical conductivity, hydrophilicity, extensive surface area, and an easily tunable structure [16]. Especially the delaminated Ti₃C₂ MXene, whose exposed surface is susceptible to functionalization by termination groups (T_x), e.g., -O, -OH, and -F, has a strong electrostatic attraction to metals and metal oxides [17]. Therefore, applying MXene to the design of electrocatalysts becomes an attractive idea. In the density functional theory (DFT) study of Yang et al. [18], Mo₂VC₂-MXene was able to synthesize urea through spontaneous co-adsorption of CO₂ and N₂. They found that Mo₂VC₂ had excellent selectivity for promoting the conversion of *CO₂ to *CO. Govindan et al. [19] successfully constructed the Ru-Pd/WO₃/MXene heterostructures. The as-prepared electrocatalyst exhibited a maximum urea yield of 227 μg·h⁻¹·mg_cat.⁻¹ with a FE of 23.7% at −0.6 V vs. RHE.

Herein, we innovatively fabricated an efficient catalyst by doping copper on CeO₂/MXene nanosheets (Cu-CeO₂/MXene). Urea was synthesized by electrocatalysis through C-N coupling. The theoretical computations show that Cu-CeO₂/MXene is more favorable for the adsorption of *NO₃ intermediates than CeO₂/MXene and the Cu-doped catalyst has much lower energy barriers at the potential-determining step (PDS) of the urea synthesis reaction than the undoped CeO₂/MXene catalyst. The prepared electrocatalyst shows a yield rate of 505.1 μg·h⁻¹·mg_cat.⁻¹ with a FE of 6.3% at −0.8 V vs. RHE. The results are superior to some of the catalysts that have been reported to date (Table S1). Our work provides a novel idea for designing catalysts for electrochemical synthesis of urea.

2. Results and Discussion

As plotted in Figure 1a, delaminated MXene nanosheets were obtained by etching the MAX phases (M, transition metal; A, A-group element; X, C or N) using LiF/HCl as etching agents [16]. CeO₂/MXene nanosheets were prepared as catalyst substrates through wet impregnation and calcination, and Cu-CeO₂/MXene catalysts were synthesized through the same methods. The scanning electron microscopy (SEM) image (Figure 1b) of MXene clearly demonstrates the two-dimensional nanosheet morphology. It indicates that MXene nanosheets have a large surface area that facilitates the anchoring and dispersion of nanoparticles. As seen from Figure 1c,d, the Cu-doped CeO₂ nanoparticles are uniformly deposited on the MXene surfaces. Furthermore, the high-resolution transmission electron microscopy (HRTEM) picture of Cu-CeO₂/MXene (Figure 1e) shows the lattice fringes spaced at 0.311, 0.275, and 0.234 nm, indexed to the (111), (200) planes of CeO₂ and the (103) plane of MXene [15,20,21]. It confirms the successful introduction of CeO₂ onto MXene nanosheets.

Figure 2a illustrates the X-ray diffraction (XRD) images of the prepared samples. The top curve reveals the XRD image of MXene. Compared to the incompletely etched Ti₃AlC₂ MAX (Figure S1), the characteristic peaks such as peaks at ~9° and ~39° completely disappear, while the peaks corresponding to Ti₃C₂ MXene appear [22]. This proves that the Al in the MAX phase has been completely etched out. The following two curves in Figure 2a show that both samples exhibit characteristic peaks of CeO₂ (PDF#34-0394), demonstrating that Cu doping doesn’t affect the crystal structure of CeO₂. And there are no characteristic peaks observed associated with crystalline Cu, suggesting that the Cu is successfully doped [23]. Furthermore, the signal of MXene is not observed in the two curves, which may be due to the low MXene content in the composites. The X-ray photoelectron spectroscopy (XPS) tests were applied to investigate the chemical composition of Cu-CeO₂/MXene. For the Ce 3d spectrum (Figure 2b), the peak of Ce 3d_5/2 at 882.3 eV and Ce 3d_3/2 peak at 898.1 eV correspond to Ce⁴⁺ species [24,25]. The Ce 3d_5/2 peak at 884.6 eV and Ce 3d_3/2 peak at 900.6 eV come from Ce³⁺. The Cu 2p spectrum (Figure 2c) exhibits the peaks of 933.5 eV and 953.5 eV that are only indexed to Cu²⁺, suggesting that the main copper species are Cu²⁺ [26]. Furthermore, the spectrum of O 1s in Figure 2d demonstrates three peaks. The peak at 529.4 eV corresponds to the lattice O from Ce-O and Cu-O, the peak centered at 530.9 eV is attributed to V_o while the peak of 533.4 eV comes from adsorbed O [27]. The above results indicate that Cu has been successfully doped without aggregation. The exact amount of Cu elements in Cu-CeO₂/MXene is further characterized by XPS and their atomic percentage is 5.59%.

Subsequently, the electrocatalytic performance of Cu-CeO₂/MXene, Cu-CeO₂, and CeO₂/MXene were evaluated and compared using an H-cell. A total of 0.1 M KHCO₃, which contained 50 mM KNO₃ acted as the electrolyte and CO₂ was continuously bubbled as feeding gas. The linear sweep voltammetry (LSV) curves of Cu-CeO₂/MXene, Cu-CeO₂, and CeO₂/MXene in Ar/CO₂-saturated electrolytes are plotted in Figure 3a and Figure S2. It can be seen that the current density in CO₂-saturated electrolytes is significantly higher than that in Ar-saturated electrolytes and the current density of Cu-CeO₂/MXene is higher than that of CeO₂/MXene, suggesting the superior electrocatalytic activity of Cu-CeO₂/MXene induced by Cu-doping towards the urea electrocatalytic synthesis from CO₂ and NO₃⁻.

The chronoamperometry tests were performed at different working potentials to investigate the electrocatalytic activity of catalysts. The urea was quantified via urease decomposition and chromogenic methods based on ultraviolet and visible (UV–vis) absorption spectra (Figure S3) [28,29]. The corresponding calibration curve is plotted in Figure S4. As Figure 3b,c shown, Cu-CeO₂/MXene exhibits significantly superior urea yield rate and FE compared to Cu-CeO₂ and CeO₂/MXene. Meanwhile, the contents of MXene and the amount of Cu doping are regulated. As shown in Figure 3d,e, the catalyst exhibits the highest urea production performance when the mass percentage of MXene is 10 wt% as well as when the mass percentage of Cu is 10 wt%. The maximum yield rate of urea can reach 505.1 μg·h⁻¹·mg_cat.⁻¹ with a FE of 6.3% at −0.8 V vs. RHE. Cu-CeO₂/MXene exhibits superior urea synthesis properties to some of the reported catalysts (Table S1). Using the chromogenic method to analyze the by-products, such as NH₃ and NO₂⁻ [30], the calibration curves are plotted in Figures S4 and S5, respectively. The remaining by-products, e.g., H₂ and CO, were detected through gas chromatography. Figure 3f shows the possible products of the electrosynthesis urea, including H₂, CO, urea, NO₂⁻, and NH₃. It can be seen that when the reaction potential surpasses −0.8 V vs. RHE, competitive reactions, e.g., hydrogen evolution reaction (HER) and NO₃⁻ reduction reactions (NO₃⁻RR) intensify [31,32]. The competitive reactions for the generation of by-products lead to a decrease in the selectivity of urea synthesis. Therefore, −0.8 V vs. RHE is the optimum potential for urea synthesis.

To further compare the activity differences of these three electrocatalysts, we tested the electrochemical active surface area (ECSA) through double-layer capacitance (C_dl) measurement [33]. As a result (Figure S6), Cu-CeO₂/MXene has a larger ECSA than both Cu-CeO₂ and CeO₂/MXene. Furthermore, electrochemical impedance spectroscopy (EIS) of the above three materials were analyzed. As shown in Figure S7, both Cu-CeO₂/MXene and Cu-CeO₂ exhibite significantly smaller semicircle radii than CeO₂/MXene. Table S2 indicates that the charge transfer resistance (R_ct) values of Cu-CeO₂/MXene and Cu-CeO₂ are 7.483 and 6.975 Ω, respectively, much lower than that of CeO₂/MXene (10.59 Ω). And the slope in the high-frequency region of Cu-CeO₂/MXene is more vertical than those of Cu-CeO₂ and CeO₂/MXene. It corresponds to the fact that Cu-CeO₂/MXene has the lowest Warburg impedance (4.665 Ω) compared to Cu-CeO₂ (10.38 Ω) and CeO₂/MXene (8.021 Ω) as Table S2 showed. These results show that Cu-CeO₂/MXene has lower impedances and faster diffusion rate [34].

Furthermore, the stability of Cu-CeO₂/MXene was evaluated at −0.8 V vs. RHE. As shown in Figure 4a, after 6 successive cycles, the yield rate and FE of urea hardly decreased. Also, there was no significant decrease in current density over 30 h of constant potential electrolysis (Figure 4b). These confirm the excellent electrocatalytic stability of Cu-CeO₂/MXene. The crystal structure of Cu-CeO₂/MXene after electrolysis was analyzed through XRD (Figure S8). The characteristic peaks are well maintained after 30 h of electrolysis, proving the good stability of the catalyst.

Next, three controlled experiments on Cu-CeO₂/MXene were carried out to exclude the possible interferences in the electrosynthesis of urea: (i) chronoamperometry at open circuit potential (OCP) in CO₂-saturated 0.1 M KHCO₃ that contained 50 mM KNO₃ for 0.5 h; (ii) chronoamperometry at −0.8 V vs. RHE in Ar-saturated 50 mM KNO₃ for 0.5 h; (iii) chronoamperometry at −0.8 V vs. RHE in CO₂-saturated 0.1 M KHCO₃ for 0.5 h. We compared the NH₃ absorbance of the electrolyte in experiment (i) with that of blank electrolyte via UV–vis absorption spectra and there was no significant difference between them (Figure S9). The UV–vis absorption spectra in both cases (ii) and (iii) (Figure S10) indicate that only trace amounts of urea are detected (Figure 4c). In summary, it is demonstrated that urea can only be detected when NO₃⁻ and CO₂ coexist, suggesting that the urea electrosynthesis on Cu-CeO₂/MXene does indeed result from the C-N coupling between them. Meanwhile, in order to study the nitrogen source of the produced urea, we conducted ¹⁵N isotope tracer experiments with ¹⁵NO₃⁻. The ¹H NMR spectrum was presented in Figure 4d. In contrast to three peaks presented by ¹⁴NH₄⁺ from ¹⁴NO₃⁻ [35], the ¹H NMR spectrum of ¹⁵NH₄⁺ shows two peaks when using ¹⁵NO₃⁻, which further confirms that urea originates from the C-N coupling of CO₂ and NO₃⁻.

Finally, we used DFT calculations to further explore the mechanism of C-N coupling during the electrocatalytic synthesis of urea. The specific structures of Cu-CeO₂/MXene and CeO₂/MXene were constructed to evaluate their urea synthesis property (Figure S11). As shown in Figure 5a, CeO₂/MXene has a weaker adsorption capacity for the initial intermediate *NO₃ with an adsorption energy ΔE_ad = −0.303 eV. While Cu doping provides adsorption sites for the catalyst, the *NO₃ intermediates chemisorb on the Cu sites with a strong ΔE_ad of −2.742 eV. The charge density difference of *NO₃ adsorption on the surfaces of the two catalysts are shown in Figure 5b. Cu-CeO₂/MXene possesses more intensive charge transfer than CeO₂/MXene and the charge is enriched at the Cu-O bond. CeO₂/MXene requires more energy to make the *NO₃ intermediate bind to the catalyst surface. Thus, *NO₃ intermediate is preferentially adsorbed and activated on Cu-CeO₂/MXene surface. For the *CO intermediate, another key intermediate of the C-N coupling, Cu-CeO₂/MXene also exhibits stronger adsorption energy (ΔE_ad= −1.266 eV) than CeO₂/MXene (ΔE_ad = −0.154 eV) as Figure S12 shows. Figure 5c illustrates the partial density of states (PDOS) images and d-band centers (E_d) of Ce 3d orbits for Cu-CeO₂/MXene and CeO₂/MXene. The E_d of the Ce atom of Cu-CeO₂/MXene is −1.701 eV, compared with CeO₂/MXene (E_d = −2.365 eV), shifting more positively towards the Fermi level (E_f = 0 eV). This implies that the interaction between intermediates and active sites is stronger in Cu-CeO₂/MXene than in CeO₂/MXene [36]. Therefore, Cu doping contributes to the adsorption of reactants on the catalyst in our system.

Moreover, the free energy changes in the urea synthesis process were computed (Figure 5d). Herein, CeO₂/MXene has several energy barriers in the reaction pathway [37,38,39]. Among them, the adsorption of *NO₃ intermediate is the PDS whose energy barrier is 1.924 eV, whereas the free energy shows a decreasing trend in this step of Cu-CeO₂/MXene. It is consistent with the result of adsorption energy comparison. Compared to CeO₂/MXene (ΔG = 1.924 eV), Cu-CeO₂/MXene possesses a lower reaction energy barrier (ΔG = 0.841 eV, *NO₃ → *NO₃H) throughout the reaction process. In particular, the second C-N coupling (*NOCO → *NOCONO) of CeO₂/MXene requires an energy barrier (0.597 eV), while the two C-N coupling steps of Cu-CeO₂/MXene are both thermodynamically spontaneous. In addition, the HER competition reaction during the synthesis of urea was investigated. As shown in Figure S13, the doping of Cu is able to suppress hydrogen production with a greater energy barrier over the CeO₂/MXene surface. Therefore, the doping of Cu on CeO₂ substrate leads to a reduction in the competitive reaction. To sum up, the doping of Cu activates the Ce sites and forms Cu-Ce sites, which optimize the adsorption capacity for the key intermediates and inhibit hydrogen production. At the same time, the reaction energy barriers for urea synthesis are lowered, thus improving the urea synthesis performance.

3. Experimental Section

3.1. Synthesis of MXene Nanosheets

Ti₃C₂T_x MXene nanosheets were produced by etching aluminum atoms from the Ti₃AlC₂ MAX phase [40,41]. A total of 1.6 g of LiF was mixed with 20 mL of 12 M HCl in a 30 mL Teflonlined vessel, and this mixture solution was stirred in a 60 °C oil bath for 20 min. After that, 1.0 g of Ti₃AlC₂ MAX was added into the vessel slowly to avert rapid exotherm, and stirred continuously in the 60 °C oil bath for 48 h. The etched mixed liquid was then centrifuged to obtain multilayered Ti₃C₂T_x MXene. The obtained precipitate was washed several times by 2M HCl to remove excess LiF. Then, this precipitate was washed repeatedly by centrifugation with deionised water until a black viscous liquid appeared in the upper layer. The pH value of the supernatant was about 6. At this point, the bottom sediment would exhibit significant swelling. It was then centrifuged 5 times without pouring it off. Each of the above centrifugations was set to 5000 rpm for 2 min. After centrifuging, multilayered Ti₃C₂T_x was homogeneously dispersed by ultrasonic shaking in a cold water bath for 20 min. At last, the sample was centrifuged at 3500 rpm for 20 min to receive delaminated MXene. The resulting liquid was then rapidly frozen with liquid nitrogen and subsequently freeze-dried.

3.2. Synthesis of CeO₂/MXene

Different weights of MXene nanosheets (16, 32 and 52 mg) were dispersed in 10 mL of deionized water. A total of 0.72 g of Ce(NO₃)₃·6H₂O was added to the dispersion, stirring for 2 h. Subsequently, the mixed liquid is freeze-dried for 24 h. After freeze-drying, the powdered material obtained was transferred to a tube furnace, annealing at 600 °C Ar atmosphere for 3 h. The heating rate is 5 °C/min. Then, yellowish CeO₂/MXene was prepared. CeO₂/MXene samples that contained different amounts (16, 32 and 52 mg) of MXene were marked as CeO₂/MXene—5%, CeO₂/MXene—10% and CeO₂/MXene—15%, respectively.

3.3. Synthesis of Cu-CeO₂/MXene

50 mg of prepared CeO₂/MXene was dispersed in 10 mL of deionized water, ultrasonicating for 15 min. Different volumes (1.42, 2.99, and 4.74 mL) of 5.0 mg·mL⁻¹ CuCl₂·2H₂O aqueous solution were added and stirried for 20 min. A total of 5 mL of Na₂CO₃ solution (50 mg·mL⁻¹) was added and stirried for 2 h. After the reaction, wash the precipitate thoroughly with deionized water several times and freeze-dry it. Prepared powders were then calcinated in tube furnace at Ar atmosphere at 250 °C for 2 h. The heating rate is 5 °C/min. Cu-CeO₂/MXene samples containing different amounts (1.42 mL, 2.99 mL and 4.74 mL) of CuCl₂·2H₂O aqueous solution (5.0 mg·mL⁻¹) were labelled as 5%—Cu-CeO₂/MXene, 10%—Cu-CeO₂/MXene, and 15%—Cu-CeO₂/MXene, respectively.

3.4. Characterization of Catalysts

The morphologies of electrocatalysts were characterized via field emission scanning electron microscopy (SEM, ZEISS VLTRA-55, Carl Zeiss, Oberkochen, German) and transmission electron microscopy (TEM, Tecnai G2 F20, FEI, Hillsboro, OR, USA). X-ray diffraction (XRD) tests were performed by D8 ADVANCE X-ray diffractometer (Bruker, Billerica, MA, USA). X-ray photoelectron spectroscopy (XPS) measurements were investigated via ESCALAB 250 Xi X-ray photoelectron spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). The gas phase products of the catalytic reactions were quantified by GC9790 plus (FULI instruments, Wenling, China). ¹H nuclear magnetic resonance (NMR) tests were investigated by Ascend 600 MHz NMR spectrometer (Bruker, Billerica, MA, USA).

3.5. Electrochemical Measurements

All electrochemical tests were performed in a H-cell with a three-electrode configuration using a CHI 760F workstation, CH Instruments, Inc., Houston, TX, USA. A Nafion 117 membrane (Dupont, Wilmington, DE, USA) was pretreated for separating the H-cell. Prepared electrocatalyst was loaded on a carbon paper (Hesen, Guangzhou, China) to act as the working electrode. The counter electrode was a carbon rod and the reference electrode was a Ag/AgCl filled with saturated KCl solution. A total of 2 mg of electrocatalyst was dispersed in the catalyst ink that was made with deionized water (475 μL), isopropanol (475 μL), and 5 wt% Nafion (50 μL) with ulatrasonication of 40 min. Then, 50 μL of homogenous catalyst ink was taken and dropped on half of the carbon paper, the geometric area of which was 2 × 1 cm⁻².

Before electrocatalytic synthesis, CO₂ (99.999%, flow rate: 50 sccm) was pumped into the electrolyte for 30 min to degas and saturate with CO₂. When starting electrochemical measurements, the flow rate was reduced to 30 and lasted for 0.5 h. After that, the cathode electrolyte was collected to be further analyzed. In our work, the working potentials were exchanged to the RHE scale by following equation:

E(vs. RHE) = E(vs. Ag/AgCl) + 0.0591 × pH + 0.197.

3.6. Quantification of NH₃ and Urea

The generated NH₃ was quantified via the indophenol blue method and UV–vis spectrophotometry [28]. A total of 2.0 mL of 1 M NaOH (5 wt% sodium citrate, 5 wt% salicylic acid), 1.0 mL of 0.05 M sodium hypochlorite, and 0.2 mL of 1 wt% sodium nitroprusside dihydrate were added sequentially into 2.0 mL electrolyte. The solution was subsequently placed in darkness for 2 h. The absorbance from 550 to 750 nm was measured via UV–vis spectrophotometry. NH₃ concentration was calculated via the calibration curve, which was according to the correlation between concentration and absorbance of NH₃ at 662 nm.

The urea concentration was quantified via the urease decomposition. A total of 0.2 mL of 5 mg·mL⁻¹ urease was mixed with 1.8 mL electrolyte, hydrolyzing for 1 h at 37 °C. One urea molecule was hydrolyzed to one CO₂ and two NH₃. The quantification of the hydrolyzed NH₃ concentration was consistent with the method described above. The concentration of urea was calculated due to the stoichiometric conversion.

The mole concentration of urea (c_urea) was obtained by the following equation:

$${\mathrm{c}}_{\mathrm{u}\mathrm{r}\mathrm{e}\mathrm{a}}={\displaystyle \frac{60.06({\mathrm{c}}_2-{\mathrm{c}}_1)}{2\times 17}}$$

where c₂ and c₁ are the NH₃ concentrations after and before the decomposition of urease, respectively.

The urea yield rate could be obtained as follows:

$$\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}={\displaystyle \frac{\mathrm{c}\mathrm{v}}{\mathrm{t}\mathrm{m}}}$$

where c represents the urea concentration (mg·mL⁻¹); v represents the electrolyte volume (mL): t for the electrolysis time (h); m is catalyst loading (mg).

The FE of urea could be calculated via the equation below:

$$\mathrm{F}\mathrm{E}={\displaystyle \frac{\mathrm{n}\mathrm{F}\mathrm{c}\mathrm{v}}{60.06\mathrm{Q}}}\times 100\mathrm{\%}$$

where n represents the number of electron transfer, which is 16, F is the Faraday constant (96,485.3 C·mol⁻¹), c is the urea concentration, v for electrolyte volume (mL), Q is the electric quantity.

The FE of NH₃ was calculated as followed:

$$\mathrm{F}\mathrm{E}={\displaystyle \frac{\mathrm{n}\mathrm{F}\mathrm{c}\mathrm{v}}{17.03\mathrm{Q}}}\times 100\mathrm{\%}$$

where n stands for the number of electrons transferred, which is 8, F is the Faraday constant (96,485.3 C·mol⁻¹), c is NH₃ concentration, v is electrolyte volume (mL), and Q denotes the electric quantity.

3.7. Quantification of CO and H₂

The gaseous products were quantified using the gas chromatograph. H₂ was identified through the thermal conductivity detector (TCD), while CO was identified through the flame ionization detector (FID). The FE of H₂ and CO were calculated via the equation below:

$$\mathrm{F}\mathrm{E}={\displaystyle \frac{\mathrm{n}\mathrm{F}{\mathrm{S}}_1\mathrm{C}\mathrm{P}\mathrm{V}}{{\mathrm{S}}_2\mathrm{I}\mathrm{R}\mathrm{T}}}\times 100\mathrm{\%}$$

where n denotes the number of electron transfers, which is 2, F represents the Faraday constant (96,485.3 C·mol⁻¹), the S₁ stands for the product peak area and the C for the standard gas product concentration, P is a standard atmospheric pressure (101,325 Pa), V stands for the gas flow rate of CO₂ (mL·min⁻¹). S₂ is the standard gaseous product peak area, I is the total current, R is 8.314 J·mol⁻¹·K⁻¹, and T is 298.15 K.

3.8. Quantification of NO₂⁻

First, to obtain the coloring agent, 8.0 g of sulfanilamide, 0.4 g of N-(1-Naphthyl)ethylenediamine dihydrochloride, and 20 mL of phosphoric acid (r = 1.685 g·mL⁻¹) were added into 100 mL of deionized water and well-mixed. A total of 0.5 mL of electrolyte was diluted to 5 mL. A total of 0.1 mL of coloring agent was taken to mix with the 5 mL diluted electrolyte. After mixing thoroughly and reacting for 20 min, the absorbance from 440 to 640 nm was detected via UV–Vis spectrophotometer. NO₂⁻ concentration was calculated due to the calibration curve, which was according to the correlation between NO₂⁻ concentration and the absorbance at 540 nm. The FE of NO₂⁻ is calculated as follows:

$$\mathrm{F}\mathrm{E}={\displaystyle \frac{\mathrm{n}\mathrm{F}\mathrm{c}\mathrm{v}}{46\mathrm{Q}}}\times 100\mathrm{\%}$$

where n represents the number of electron transfer which is 2, F represents the Faraday constant (96,485.3 C·mol⁻¹), c stands for the NO₂⁻ concentration, v stands for electrolyte volume, and Q represents electric quantity.

3.9. ECSA Measurements

The ECSA of the catalysts was assessed by C_dl measurement using a CHI 760F workstation. The CV curves in the non-Faradic region (from −0.5 to −0.3 V vs. RHE) at scan rates of 50, 60, 70, 80, 90, and 100 mV·s⁻¹ were recorded. Then, the two current values of each CV curve in the middle of the region were taken (i.e., at −0.4 V vs. RHE), and the sum of their absolute values was averaged. The obtained values were used as the Y scales. The values of scan rates were used as the X scales. The slope of the resultant line corresponded to the value of C_dl.

3.10. EIS Measurements

First, the OCP was measured by CHI 760F workstation as the initial potential for the EIS measurements. Then, the frequency range was set from 0.01 to 10⁵ Hz and the amplitude was set to 5 mV. The obtained data were fitted by Zview 3.1 software.

3.11. NMR Measurements

The ¹⁴N and ¹⁵N isotope labelling experiments were used to detect the N source in obtained urea. First, the produced urea was hydrolyzed to NH₄⁺ by urease. Then, 50 μL of HCl was added into 500 μL of electrolyte and 100 μL of DMSO-d6 was added as the deuterated solvent. The final result was the accumulation of 64 scans using a 600 MHz NMR spectrometer.

3.12. Theoretical Calculations

The calculations were obtained via the Vienna ab initio simulation program. For the treatment of electronic exchange and correlation effects, the Perdew-Burke-Ernzerhof (PBE) functional, rooted in the generalized gradient approximation (GGA), was employed. The cut-off energy of the plane-wave basis set was 550 eV for energy convergence. The Brillouin zone was sampled with a 3 × 3 × 1 k-point to achieve sufficient precision. A vacuum region of 17 Å was created at the top of the model to minimize spurious interactions between periodic cells. The convergence threshold for electronic self-consistency was 10⁻⁵ eV. The convergence threshold for ionic relaxation was set at 0.02 eV/Å.

Free energy change (ΔG) in each adsorbed intermediate was computed using the following expression:

ΔG = ΔE + ΔE_ZPE – TS

where ΔE stands for changes in electron energy, ΔE_ZPE represents zero-point energy, S for the entropy associated with the adsorption of intermediates. The thermodynamic corrections at the reaction temperature (298 K) were calculated using the VASPKIT 1.4.1 software suite.

4. Conclusions

In summary, Cu-CeO₂/MXene is evidenced as an effective catalyst that is capable of electrocatalytic synthesizing urea under ambient conditions. Large amounts of CeO₂ are firmly anchored on the MXene two-dimensional surface to form nanoparticles, which have excellent charge transfer capability. The doped-in Cu and Ce form the Cu-Ce sites that adsorb and activate the intermediates and facilitates the electrocatalytic kinetics. The prepared Cu-CeO₂/MXene achieves the C-N coupling between CO₂ and NO₃⁻ to electrocatalytically synthesize urea. Compared with undoped CeO₂/MXene, Cu-CeO₂/MXene exhibits a urea yield rate of 505.1 μg·h⁻¹·mg_cat.⁻¹ with a FE of 6.3% at −0.8 V vs. RHE. Theoretical calculations demonstrate that Cu doping is capable of introducing reactive sites for the reaction, which enhances the ability to adsorb reactants. Coordination of Cu with Ce sites lowers the energy barrier and facilitates the hydrogenation process. Our study provides some fresh ideas for electrocatalytic synthesis of urea via C-N coupling, which may be useful for the construction of other advanced electrocatalysts.