Defect-engineered N-doped carbon stabilizes Cu⁺ active sites for bifunctional CO₂ electroreduction to CO and formate

Abstract

The development of bifunctional electrocatalysts capable of steering CO₂ reduction toward selective C₁ products under mild conditions remains central to advancing next-generation electrochemical technologies. Here, we demonstrate that stabilization of Cu⁺ species by N-doped carbon derived from tea leaves (TL9) enables highly selective and durable CO₂ electroreduction to CO and formate. Uniformly dispersed Cu₂O nanoparticles supported on TL9 exhibit strong metal–support interactions and form stable Cu–N_x coordination that preserves the active Cu⁺/Cu⁰ interface during operation. Structural, spectroscopic, and electrochemical analyses reveal that this tailored interface suppresses Cu agglomeration and hydrogen evolution, promoting efficient two-electron transfer pathways. The optimized TL9/Cu-40% catalyst achieves faradaic efficiencies approaching 90% for CO and formate at −0.6 V vs. RHE and maintains over 60% selectivity after 24 h of continuous operation. These findings highlight how defect-engineered carbon supports can precisely regulate Cu oxidation states to enhance efficiency, selectivity, and stability—offering a robust design principle for bifunctional catalysts that couple renewable electricity with CO₂ valorization.

---

1. Introduction

The electrochemical reduction of carbon dioxide (CO₂RR) has attracted increasing attention as a viable approach to closing the carbon cycle while generating value-added fuels and chemicals using renewable electricity. Significant progress has been achieved in developing copper-based electrocatalysts, which uniquely enable the formation of a wide spectrum of products ranging from CO and formate to multicarbon (C₂⁺) species such as ethylene and ethanol.^1,2 However, achieving high selectivity toward desired products remains a major challenge, as competing hydrogen evolution and complex multi-electron pathways often limit the overall reaction efficiency.^3,4

Among the various CO₂RR products, CO and formate stand out as particularly attractive due to their simple two-electron transfer pathways. In contrast to C₂⁺ products that require high overpotentials, intricate C–C coupling steps, and often suffer from poor durability, the generation of CO and formate is both kinetically and thermodynamically more favorable. Consequently, these reactions can deliver higher faradaic efficiencies (FE) under milder operating conditions, providing a practical route for sustainable CO₂ utilization.⁵ Beyond efficiency, these C₁ products serve as versatile intermediates: CO is a cornerstone of syngas chemistry and an essential feedstock for downstream processes such as Fischer–Tropsch and methanol synthesis,^6–8 while formate is a stable liquid fuel that can be used directly in formate fuel cells or upgraded into formic acid and oxalate-derived polymer feedstocks.⁵ Formate, in particular, should be regarded not merely as a terminal product but as a key gateway intermediate. Its coupling to oxalate unlocks pathways toward a spectrum of downstream chemicals—including glycolic acid, ethylene glycol, glyoxal, and metal oxalates—offering a more practical and scalable alternative to the direct but less selective C₂⁺ formation route. Together, CO and formate represent a complementary product pair, balancing gaseous intermediates for large-scale synthesis with liquid fuels and chemical precursors that integrate seamlessly into existing infrastructures.

The ability of copper to toggle between Cu⁺ and Cu⁰ oxidation states is widely recognized as a crucial factor governing CO₂RR selectivity. Cu⁺ species stabilize key intermediates (e.g., *COOH for CO, *OCHO for formate), whereas Cu⁰ facilitates electron transfer and product desorption.^9–11 However, under strongly reducing potentials, Cu⁺ tends to convert irreversibly to metallic Cu.¹² One promising strategy to overcome this limitation is the use of N-doped carbon supports, which can anchor Cu⁺ species at defect sites and stabilize them via coordination with pyridinic or pyrrolic nitrogen functionalities.¹³ By tuning the metal–support interaction and dispersion, such hybrid catalysts can maintain a favorable Cu⁺/Cu⁰ balance during operation, thereby enhancing both selectivity and stability. Previous studies have demonstrated that N-doped carbon supports play a crucial role in stabilizing Cu-based active sites for selective CO₂ electroreduction. Porous N-doped carbon-supported Cu catalysts have achieved high efficiencies toward C₁ products such as CO and formate by promoting CO₂ adsorption and stabilizing Cu⁺ species.⁹ Similarly, Cu₂O/Cu-based composites immobilized on N-doped graphene have shown enhanced formate selectivity at low overpotentials through synergistic metal–support interactions that suppress hydrogen evolution and facilitate charge transfer.¹⁴ These reports collectively highlight the importance of nitrogen-containing carbon supports in regulating Cu oxidation states and reaction pathways.

In this study, we systematically investigate Cu₂O nanoparticles supported on N-doped carbon derived from tea leaves and compare their CO₂RR performance with those supported on commercial activated carbon. We demonstrate that nitrogen functionalities not only promote uniform dispersion of Cu₂O but also suppress agglomeration and excessive reduction to metallic Cu, resulting in enhanced activity and long-term durability. The optimized catalyst achieves FE approaching 90% for CO and formate at −0.6 V vs. RHE and maintains over 60% selectivity after 24 h of continuous operation. These findings highlight the advantages of targeting CO and formate as practical C₁ products and provide mechanistic insights into how support chemistry governs Cu speciation and overall CO₂RR performance.

2. Experimental

2.1 Synthesis of N-doped carbon support

N-Doped carbon derived from tea leaves was synthesized following a previously reported procedure.¹⁵ Briefly, the precursor was prepared by carbonizing and activating tea leaves pretreated with potassium carbonate (K₂CO₃) at a weight ratio of 1 : 1.5. Activation was conducted in a horizontal tubular furnace at 900 °C with a heating rate of 5 °C min⁻¹ and held for 1 hour under a continuous N₂ flow of 200 cm³ min⁻¹. The resulting material was thoroughly washed with RO water and dried at 105 °C overnight. The sample was denoted as TL9, where “9” refers to the activation temperature of 900 °C.

2.2 Preparation of copper/carbon composites

Solution A was prepared by dissolving 1.11 g of copper(II) acetate monohydrate (Cu(CH₃COO)₂·H₂O) in 25 mL of distilled water and stirring at 70 °C for 10 min. Subsequently, 0.6 g of TL9 was introduced into the solution and sonicated for 10 min to ensure homogeneous dispersion. In parallel, solution B was prepared by dissolving 0.95 g of NaOH and 0.83 g of ascorbic acid in 20 mL of distilled water. Solution B was then added dropwise to solution A under continuous stirring, and the reaction was maintained for 30 min. The resulting product was collected by filtration, thoroughly washed with distilled water, and dried to obtain the TL9/Cu-x% composite, where x denotes the Cu₂O content in wt%. For comparison, YEC/Cu-40% was synthesized by replacing TL9 with commercial activated carbon (YEC-8A) under otherwise identical conditions.

2.3 Material characterizations

X-ray diffraction (XRD) patterns were obtained using a PANalytical Empyrean diffractometer with Cu Kα radiation (λ = 0.15406 nm) over a 2θ range of 10°–80° at a step size of 0.02°. The morphology and elemental composition were analyzed by scanning electron microscopy equipped with energy-dispersive X-ray spectroscopy (SEM-EDS, HITACHI SU8020). Transmission electron microscopy (TEM, HITACHI HT7700) was employed to examine the particle size and lattice fringes of Cu₂O nanoparticles. N₂ adsorption–desorption isotherms were recorded at −196 °C using a MicrotracBEL BELSORP mini-X analyzer to evaluate the textural properties. All samples were degassed under vacuum (10⁻³ kPa) at 150 °C for 6 hours prior to measurement. The specific surface area (S_BET) was calculated by the Brunauer–Emmett–Teller (BET) method. The total pore volume (V_total) was estimated from adsorption data at P/P₀ = 0.99, while the micropore volume (V_micro) was obtained using the Dubinin–Radushkevich (DR) equation. The mesopore volume (V_meso) was determined by subtracting V_micro from V_total.

X-ray photoelectron spectroscopy (XPS) was performed on a JEOL JPS-9010 MC with a Mg Kα source (12 kV, 25 mA) under a base pressure of 10⁻⁷ Pa. Thermogravimetric analysis (TGA, Mettler Toledo TGA/DSC 1) was used to quantify the Cu₂O loading in the composites. H₂ temperature-programmed reduction (H₂-TPR, BELCAT-B, BEL Japan Inc.) was conducted to evaluate the reducibility and dispersion of Cu₂O nanoparticles.¹⁶ Approximately 20 mg of sample was reduced in situ under a flow of 5% H₂/Ar from room temperature to 800 °C at a heating rate of 5 °C min⁻¹. Raman spectra were acquired using a Horiba LabRAM HR Evolution spectrometer with a 532 nm excitation laser to assess graphitization and defect density. The ratio of disordered (D) to graphitic (G) band intensities (I_D/I_G) was determined from deconvoluted peak intensities.

2.4 Electrochemical measurements

Electrochemical measurements were performed in a two-compartment, gas-tight glass H-cell separated by an anion exchange membrane (AMVN). A Pt plate (1.0 × 1.0 cm²) and a Hg/HgOH electrode in 1 M NaOH served as the counter and reference electrodes, respectively. The working electrode consisted of the catalyst coated onto carbon paper (Toray, TGP-H-120, 1.0 × 1.0 cm²).

The catalyst ink was prepared by mixing the catalyst, carbon black, and polyvinylidene fluoride (PVDF) in a weight ratio of 8 : 1 : 1 in N-methyl-2-pyrrolidone (NMP) to form a homogeneous slurry. Carbon black was added to enhance electrical conductivity and facilitate electron transport across the electrode.¹⁷ The ink was applied to the carbon paper using a doctor blade (250 µm thickness) and dried under vacuum at 80 °C overnight. The electrolyte (0.5 M KHCO₃) was continuously saturated with CO₂ at a flow rate of 20 sccm, and all experiments were conducted at room temperature.

Electrochemical impedance spectroscopy (EIS) was performed prior to CO₂RR testing to determine the solution resistance (R_s) and charge-transfer resistance (R_ct) from Nyquist plots. R_s was used to correct the iR drop, and all potentials were converted from the Hg/HgOH reference to the reversible hydrogen electrode (RHE) scale for accurate comparison of catalytic activity.

Gaseous products (CH₄, C₂H₄, and C₂H₆) were quantified using gas chromatography with a flame-ionization detector (GC-FID) equipped with a Porapak Q (60/80) column and He as the carrier gas. H₂ and CO were analyzed using GC with a thermal-conductivity detector (GC-TCD) equipped with a molecular sieve 5A column and He as the carrier gas. Liquid products were analyzed by high-performance liquid chromatography (HPLC) equipped with a refractive-index detector (RID) using 5 mM H₂SO₄ as the mobile phase and a Shodex SH1011 Sugar column as the stationary phase. Moreover, the formation of formate was verified by ¹H NMR spectroscopy of the post-reaction electrolyte using Bruker Avance III™ HD 600 MHz, and D₂O as the solvent. All electrochemical measurements and products quantifications were performed at least three times. The reported FE represent the average values, and the error bars correspond to the standard deviation.

3. Results and discussion

3.1 Structural characteristics of the carbon samples and composites

The porous characteristics of all samples are summarized in Table S1. Both TL9 and YEC exhibit comparable BET surface areas (S_BET) and total pore volumes. Their N₂ adsorption–desorption isotherms (Fig. 1a) display combined type I and type IV profiles with H4-type hysteresis loops, indicative of predominantly microporous structures containing a small fraction of slit-like mesopores. The slightly more pronounced hysteresis in TL9 suggests a marginally higher mesopore content than in YEC, consistent with the V_meso values in Table S1. However, this difference is relatively minor and unlikely to significantly affect their comparative electrocatalytic performance. As shown in Fig. 1b, both samples possess similar pore size distributions dominated by micropores (∼1 nm).

Fig. 1 (a) N₂ adsorption–desorption isotherms and (b) NLDFT pore size distribution of TL9 and YEC samples. (c) N₂ adsorption–desorption isotherms and (d) NLDFT pore size distribution of TL9/Cu-40% and YEC/Cu-40% composites.

Upon Cu₂O incorporation, the adsorbed volume and corresponding surface area and pore volume decreased progressively with increasing Cu₂O loading (Fig. S1a). Notably, TL9/Cu-50% exhibits a type II isotherm (Fig. S1b), characteristic of macroporous adsorbents with strong adsorbate–adsorbent interactions,¹⁸ accompanied by a substantial loss of S_BET and V_micro. This behavior indicates that excessive Cu₂O loading likely blocks or fills the available pores of the carbon support. Fig. 1c compares the isotherms of TL9/Cu-40% and YEC/Cu-40%, both showing combined type I/IV features with H4-type hysteresis loops. The depletion of micropores (∼1 nm) observed for TL9/Cu-40% (Fig. 1d) suggests partial blockage of micropores during Cu₂O deposition, whereas the reduction in pore volume of YEC around 1.5 nm implies Cu₂O deposition within larger mesopores or interparticle voids.

Raman spectra of TL9, YEC, TL9/Cu-40%, and YEC/Cu-40% (Fig. 2) exhibit two prominent peaks at ∼1340 cm⁻¹ (D band) and ∼1580 cm⁻¹ (G band), corresponding to disordered and graphitic carbon structures, respectively.¹⁹ Deconvolution of the spectra reveals four components: D* (∼1200 cm⁻¹), D (∼1340 cm⁻¹), D** (∼1530 cm⁻¹), and G (∼1590 cm⁻¹), representing various C=C stretching and C–H wagging modes associated with structural disorder.²⁰ The degree of disorder, evaluated from the I_D/I_G ratio using the four-peak model,²¹ follows the order TL9 > YEC > TL9/Cu-40% > YEC/Cu-40% (Table S2). Specifically, TL9 (3.42) exhibits higher disorder and edge-plane exposure than YEC (2.94). After Cu₂O incorporation, I_D/I_G increases to 3.83 and 3.06 for TL9/Cu-40% and YEC/Cu-40%, respectively, indicating that Cu₂O disrupts the sp² network and introduces additional defects.²² On the other hands, N-doped defect sites—particularly pyridinic and pyrrolic N—are known to act as preferential anchoring centers for Cu-based species. During Cu₂O deposition, these pre-existing defects may be selectively occupied rather than newly generated. Hence the attachment of Cu₂O nanoparticles at these defect-rich regions perturbs the local π-conjugation and modifies the Raman scattering cross-section of neighboring bonds.²³ Because Raman spectroscopy provides a spatially averaged measure of disorder, the preferential consumption and electronic modification of defect sites can manifest as a higher apparent I_D/I_G ratio. Therefore, the observed increase likely reflects a combination of (i) perturbation of the sp² lattice at metal–support interfaces and (ii) selective deposition of Cu₂O onto intrinsic defect sites. The smaller increase of I_D/I_G ratio in YEC/Cu-40% suggests weaker Cu₂O anchoring on YEC compared to TL9.

Fig. 2 Raman spectra of (a) YEC, (b) TL9 (c) YEC/Cu-40%, and (d) TL9/Cu-40%. Note that the circle and solid lines represent the raw data and curve fitting after peak deconvolution, respectively. The peak intensity ratio, I_D/I_G, is also shown in each figure.

Thermogravimetric analysis (TGA, Fig. S2) supports the Raman findings. TL9 exhibits an earlier onset of decomposition—approximately 50 °C lower than YEC—reflecting its more defect-rich and less thermally stable framework.²⁴ Although such defects decrease structural stability, they are often catalytically beneficial.²⁵ The residual weight (∼20%) of TL9 likely arises from inherent mineral components (K, Si, Ca, Mg, P, Na) in tea-leaf-derived carbon.^26,27 These residues are not catalytically active for CO₂ reduction but can marginally lower the electrical conductivity of the carbon matrix.²⁸

X-ray diffraction (XRD) analysis was conducted to determine the crystalline phases of copper species in the composites (Fig. 3). The diffraction peaks of YEC/Cu-40% at 36.4°, 42.3°, and 61.4° correspond to the (111), (200), and (220) planes of cubic Cu₂O (PDF# 05-0667). Additional peaks at 43.3° and 50.4° correspond to metallic Cu (PDF# 04-0836), indicating partial reduction of Cu⁺ to Cu⁰ during synthesis. Such reduction is typically observed when the Cu precursor lacks sufficient anchoring functionalities.²⁹ In contrast, TL9/Cu-40% displays only Cu₂O reflections, with no detectable Cu⁰ peaks, suggesting that nitrogen functionalities in TL9 stabilize Cu⁺ species.³⁰ The stabilization effect persists across all Cu₂O loadings (10–50%), as confirmed by XRD (Fig. S3).

Fig. 3 XRD patterns corresponded to YEC/Cu-40% and TL9/Cu-40% together with the references, metallic Cu and Cu₂O.

Based on the Scherrer equation, the average Cu₂O crystallite size was ∼16 nm for both TL9/Cu-40% and YEC/Cu-40%. However, transmission electron microscopy (TEM) reveals distinct nanoparticle morphologies (Fig. 4). In YEC/Cu-40%, Cu₂O nanoparticles are agglomerated into polycrystalline structures (∼30 nm; Fig. 4b). Conversely, TL9/Cu-40% shows uniformly dispersed Cu₂O nanoparticles with an average diameter of ∼20 nm (Fig. 4a). The evolution of Cu₂O dispersion with increasing loading is shown in Fig. S4: at low loadings (10–20%), partial coalescence is observed; at 30%, secondary nucleation leads to uniform nanoparticle formation; and above 40%, agglomeration again dominates.^31,32 High-resolution TEM (Fig. 4c and d) confirms lattice fringes with d-spacing of 0.256 nm, consistent with the (111) plane of Cu₂O, with TL9/Cu-40% displaying single-crystal domains while YEC/Cu-40% forms multi-grained polycrystals.

Fig. 4 TEM images of (a) TL9/Cu-40% and (b) YEC/Cu-40% composites. The inset shows particle size distribution of each sample. High-resolution TEM images (c) and (d) highlight lattice fringes corresponding to the (111) plane of Cu₂O (d = 0.256 nm) in both samples.

Scanning electron microscopy (SEM, Fig. 5) further supports the TEM results. TL9/Cu-40% exhibits homogeneously distributed Cu₂O nanoparticles across the carbon surface, with few large aggregates (Fig. 5a), whereas YEC/Cu-40% displays large Cu₂O clusters localized in interparticle voids (Fig. 5b). Higher-magnification images reveal average particle sizes of ∼27 nm for TL9/Cu-40% and ∼35 nm for YEC/Cu-40% (Fig. 5c and d). Although SEM values are slightly larger due to imaging conditions, both analyses confirm superior dispersion and smaller particle size on TL9. The defect-rich surface of TL9 provides abundant anchoring sites, facilitating uniform nucleation and suppressing Cu₂O agglomeration.

Fig. 5 SEM images of (a) TL9/Cu-40% and (b) YEC/Cu-40% composites along with their high-resolution images (c) and (d). The inset shows particle size distribution of each sample.

The evolution of Cu₂O morphology with increasing loading on TL9 is illustrated in Fig. S5. At 10%, Cu₂O nanoparticles are well dispersed, nucleating preferentially at defect sites (Fig. S5a). As the loading increases to 20%, these sites become saturated, leading to partial coalescence (Fig. S5b). At 30%, a bimodal microstructure appears, combining agglomerated and newly nucleated nanoparticles (Fig. S5c). At 40%, both coarsened clusters and secondary dispersion coexist (Fig. S5d), whereas at 50% loading, agglomeration becomes dominant (Fig. S5e). This loading-dependent morphological evolution is consistent with nucleation–coalescence dynamics commonly observed in heterogeneous catalytic systems.^31,32

3.2 Reducibility and surface functionality

The H₂-TPR profiles of the composites exhibit three main peaks at low, moderate, and high temperatures (Fig. 6), corresponding to the stepwise reduction of Cu₂O to metallic Cu. These peaks are associated with the reduction of highly dispersed nanoparticles, loosely bound nanoparticles, and agglomerated Cu₂O particles, respectively.^16,33 For YEC/Cu-40%, the first reduction peak appears at 208 °C. In contrast, TL9/Cu-40% displays two sharp reduction peaks at 221 °C and 234 °C, which can be ascribed to uniformly distributed reducible Cu species of various particle sizes, consistent with the SEM and TEM observations. The upward shift of the first peak for TL9/Cu-40% signifies stronger metal–support interactions.³⁴ Moreover, YEC/Cu-40% exhibits an additional high-temperature peak at 275 °C, arising from the reduction of highly agglomerated Cu₂O nanoparticles, as evident in the SEM images. The broader and lower-temperature second peak at 228 °C for YEC/Cu-40% further suggests weaker Cu₂O–support interactions compared with TL9/Cu-40%. Together, the H₂-TPR profiles corroborate the microscopic evidence that TL9 effectively promotes uniform Cu₂O dispersion with stronger interfacial bonding.

Fig. 6 H₂-TPR profiles corresponded to TL9/Cu-40% and YEC/Cu-40%.

The reduction behavior of the TL9/Cu series (10–50%) shows a clear loading-dependent trend (Fig. S6). At 10% loading (Fig. S6a), two reduction peaks are observed at 231 °C and 280 °C. The lower-temperature peak reflects strong interactions between Cu₂O and defect sites on TL9,³⁴ while the higher-temperature peak likely arises from a small fraction of larger nanoparticles.^35,36 As the loading increases to 20% (Fig. S6b), both peaks intensify, indicating a greater number of reducible species. The reduction at 240 °C corresponds to loosely bound Cu₂O clusters, in agreement with TEM (Fig. S4b) and SEM (Fig. S5b) results showing partial coalescence.³² At 30–40% loadings (Fig. S6c and d), sharper and more intense reduction peaks appear, reflecting the secondary nucleation of Cu₂O nanoparticles and the predominance of uniformly dispersed species. The strong reduction signal at 224 °C for TL9/Cu-40% indicates a higher fraction of these well-dispersed particles relative to TL9/Cu-30%. Partial coalescence remains evident, as seen from the secondary peaks at 237 °C (TL9/Cu-30%) and 234 °C (TL9/Cu-40%). This behavior is consistent with previous reports for supported copper catalysts, where intermediate metal loadings favor homogeneous dispersion and narrow distributions of reducible sites.¹⁶ At 50% loading (Fig. S6e), a new peak emerges at 248 °C, indicating the formation of larger, weakly bound aggregates similar to those in YEC/Cu-40% (Fig. S6f). Overall, the 30–40% loading range yields the most uniform and reducible Cu species, whereas excessive loading promotes heterogeneity and particle agglomeration.

The XPS survey spectrum of TL9 (Fig. S7a) confirms the presence of C, N, and O as the main surface elements, with nitrogen comprising 0.43 at%. Upon Cu₂O deposition, distinct Cu 2p signals appear in TL9/Cu-40% (Fig. S7b), confirming successful Cu incorporation. The N 1s spectrum of TL9 (Fig. 7a) exhibits three main peaks at 398.0, 400.0, and 401.0 eV, attributed to pyridinic-, pyrrolic-, and graphitic-N, respectively.^37–40 After Cu₂O loading, a new peak emerges at 396.0 eV, concurrent with a marked decrease in pyrrolic-N intensity (Fig. 7b). The new feature is assigned to nitridic-like Cu–N_x species,^41–44 formed via coordination of Cu with deprotonated pyrrolic-N.⁴⁵ This coordination facilitates electron donation from N to Cu, stabilizing Cu⁺ sites and preventing over-reduction to Cu⁰. The resulting Cu–N_x bonding rationalizes the preservation of catalytically active Cu⁺/Cu⁰ species during CO₂RR.^45,46

Fig. 7 N 1s spectrum of (a) TL9 and (b) TL9/Cu-40%.

The C 1s spectra (Fig. S8) display typical peaks corresponding to C–C (284.8 eV), C–O (286.0 eV), C=O (288.0 eV), and O–C=O (289.6 eV). The O 1s spectra (Fig. S9) reveal components at 531.0 eV (O–C=O), 532.0 eV (C=O), 533.4 eV (C–O), and 535.0 eV (–OH).^47–49 A new feature at 530.5 eV appears in the composites, corresponding to Cu–O bonds in Cu₂O/CuO species, further confirming the presence of Cu₂O.

High-resolution Cu 2p spectra (Fig. 8a) show signals at ∼932–933 eV (Cu⁺/Cu⁰) and ∼934–935 eV (Cu²⁺), with shake-up satellites at 940–945 eV. Because Cu 2p cannot unambiguously distinguish Cu⁰ from Cu⁺,¹³ Cu LMM Auger analysis was performed (Fig. 8b).⁵⁰ The Auger spectra reveal that TL9/Cu-40% consists mainly of Cu⁺ (∼70%), with minor fractions of Cu⁰ (∼14%) and Cu²⁺ (∼16%). By contrast, YEC/Cu-40% contains a larger proportion of Cu⁰ (∼34%) alongside Cu⁺ (∼52%) and Cu²⁺ (∼14%), as summarized in Table S3. These findings agree well with the XRD results, where metallic Cu reflections appear only in the YEC/Cu composite. Overall, the XPS and TPR analyses confirm that the nitrogen functionalities in TL9 effectively stabilize Cu⁺ species, suppressing their reduction to Cu⁰ during synthesis.²⁹ To clarify the origin of Cu–N_x coordination despite the low overall nitrogen content, the evolution of Cu–N_x species with increasing Cu loading was quantified by XPS and is summarized in Fig. S10 and Table S4. Notably, the relationship between Cu at% and Cu–N_x content reveals a distinct saturation trend. The Cu–N_x concentration increases in proportion to Cu loading up to approximately 4.4 at% Cu (corresponding to a nominal loading of 40 wt%), beyond which it reaches a plateau despite further increases in total Cu content. This behavior indicates that nitrogen defect sites offer a finite yet sufficient number of strong anchoring sites that become fully occupied at intermediate Cu loadings. Additional Cu introduced beyond this threshold is therefore deposited predominantly as non-coordinated Cu₂O. These results highlight that effective stabilization of Cu does not require a stoichiometric N–Cu ratio; rather, a small fraction of strategically located nitrogen defects can anchor a large Cu population through localized Cu–N_x coordination.

Fig. 8 Cu 2p (left) and Cu LMM (right) spectra of (a) YEC/Cu-40% and (b) TL9/Cu-40%.

In summary, the structural, reducibility, and surface analyses collectively demonstrate that TL9 supports the uniform dispersion of Cu₂O nanoparticles strongly anchored at nitrogen defect sites, whereas YEC promotes larger aggregates with weaker metal–support interactions. The stabilization of Cu⁺ species in TL9—corroborated by XPS, Cu LMM, and H₂-TPR results—underscores the critical role of nitrogen functionalities in maintaining active Cu⁺/Cu⁰ interfaces essential for efficient CO₂ reduction catalysis.

3.3 Evaluation of the CO₂RR performances

Electrochemical impedance spectroscopy (EIS) was conducted to elucidate the interfacial kinetics and transport characteristics of the Cu-loaded carbon catalysts. Fig. 9 shows the Nyquist plots and equivalent-circuit fitting for TL9/Cu-40% and YEC/Cu-40%. Both samples exhibit a small semicircle in the high-frequency region, followed by a depressed mid-frequency response and a low-frequency Warburg tail.

Fig. 9 Nyquist plots for the EIS analysis on the TL9/Cu-40% and the YEC/Cu-40% in 0.5 M KHCO₃ solutions after purged with CO₂ for 20 min at open circuit potential (inset: equivalent circuit).

The first high-frequency semicircle is assigned to ion transport within the porous carbon matrix. This interpretation follow the previous reports,^51,52 who demonstrated that porous electrodes inherently display a high-frequency arc originating from pore-channel ion transport. In our system, the high-frequency response is described by a CPE1 with n ∼ 0.55 for TL9/Cu-40%, which indicates strongly non-ideal capacitive behavior. Such CPE behavior arises when the electrode surface is heterogeneous—typically due to features such as pore networks, surface roughness, and defects—which create a distribution of local double-layer capacitances instead of a single uniform capacitor.⁵³ Therefore, the CPE in this region reflects the defect-rich and highly porous structure of the TL9 carbon support. The corresponding resistance shown in Table S5 (R_p = 3.43 Ω for TL9/Cu-40% and 2.80 Ω for YEC/Cu-40%), hence reflects pore/film resistance.

The mid-frequency region represents the actual faradaic charge-transfer resistance (R_ct) associated with the Cu active sites. Although the semicircle is depressed due to constant-phase behavior and partial overlap with diffusion, equivalent-circuit fitting allows reliable extraction of R_ct. TL9/Cu-40% shows a significantly lower charge-transfer resistance (R_ct = 26 Ω) compared to YEC/Cu-40% (R_ct = 40 Ω), indicating more favorable electron-transfer kinetics. The CPE2 exponent for this branch differs markedly between the two catalysts: TL9/Cu-40% exhibits almost ideal capacitive behavior (n ∼ 1.05), while YEC/Cu-40% shows strong interfacial heterogeneity (n ∼ 0.60). These features suggest that the Cu–electrolyte interface in TL9 is more uniform, likely due to better Cu dispersion.

At low frequencies, both catalysts display a Warburg diffusion tail, but the magnitude is substantially larger for YEC/Cu-40%, implying more severe mass-transport limitations within its carbon structure. In contrast, the lower diffusion impedance observed for TL9/Cu-40% indicates more efficient CO₂ and ion transport through its more open pore network.

Overall, the EIS analysis demonstrates that TL9/Cu-40% possesses (i) lower charge-transfer resistance, (ii) more ideal double-layer characteristics, and (iii) reduced diffusion impedance compared to YEC/Cu-40%. These features collectively explain the superior CO₂ reduction activity of the TL9-derived catalyst.

The cyclic voltammetry (CV) profiles (Fig. 10a) show that TL9/Cu-40% exhibits a slightly more negative onset potential (−0.8 V vs. RHE) than YEC/Cu-40% (−0.6 V vs. RHE), suggesting partial suppression of the hydrogen evolution reaction (HER). HER is primarily governed by surface electronic structure and the availability of metallic Cu⁰ and uncoordinated sites. YEC/Cu-40%, which contains a higher fraction of metallic Cu and agglomerated nanoparticles, facilitates rapid H* adsorption and therefore exhibits higher HER currents at a given potential. In contrast, TL9/Cu-40% maintains a higher Cu⁺/Cu⁰ ratio stabilized by Cu–N_x coordination, which kinetically suppresses hydrogen evolution despite a slightly more negative onset potential.^54,55 This is further supported by electrochemical double-layer capacitance (C_dl) measurements (Fig. 10b), where TL9/Cu-40% displays a higher capacitance (72.2 mF cm⁻²) than YEC/Cu-40% (51.9 mF cm⁻²), indicative of a larger electrochemically active surface area (ECSA). Despite the slightly delayed onset, TL9/Cu-40% demonstrates markedly superior CO₂RR activity, achieving ∼90% FE toward C₁ products at −0.6 V vs. RHE, compared with only 28% for YEC/Cu-40% (Fig. 10c and d). Remarkably, even at −0.8 V—where HER typically dominates—the FE of TL9/Cu-40% remains above 50%, underscoring its strong CO₂RR selectivity.

Fig. 10 (a) Cyclic voltammograms and (b) double-layer capacitance (C_dl) of TL9/Cu-40% and YEC/Cu-40%. FE of (c) TL9/Cu-40% and (d) YEC/Cu-40% using H-type cell in 0.5 M KHCO₃ electrolyte saturated with CO₂ (pH = 7.4).

During CO₂ electroreduction, the Cu oxidation state dynamically evolves toward a mixed Cu⁺/Cu⁰ interface.^30,56 Cu⁺ sites stabilize key intermediates such as *COOH (for CO) and *OCHO (for formate), whereas neighboring Cu⁰ domains promote electron transfer, product desorption, and C–C coupling toward C₂⁺ products.^13,56,57 Unstabilized Cu⁺ species tend to reduce to Cu⁰, forming so-called oxide-derived Cu (OD-Cu), which mainly produces CO through the *COOH pathway.¹² Nitrogen species may also act as HER sites,⁵⁸ thus competing with CO₂RR. XPS analysis confirmed that TL9/Cu-40% retains a larger Cu⁺ fraction than YEC/Cu-40%, consistent with the stabilizing effect of N dopants that anchor Cu⁺ and suppress its full reduction to metallic Cu. This stabilization preserves a favorable Cu⁺/Cu⁰ balance during operation, optimizing C₁-product formation.¹ The persistence of Cu⁺ at N-anchored sites promotes selective CO and formate generation,⁵⁷ whereas the dominance of metallic Cu and exposed N sites in YEC/Cu-40% results in lower CO₂RR selectivity and CO as the major product. These results confirm that the Cu₂O–N-carbon interaction critically governs the reaction pathway by sustaining catalytically active Cu⁺/Cu⁰ interfaces.⁵⁹

The superior CO₂RR activity of TL9/Cu-40% arises from the uniform dispersion and strong anchoring of Cu₂O nanoparticles at nitrogen defect sites. Homogeneously distributed Cu species offer a higher density of electronically consistent active centers, reflected by the larger C_dl (72.2 mF cm⁻² vs. 51.9 mF cm⁻² for YEC/Cu-40%), and thus a greater ECSA.^60,61 These features facilitate the stabilization of *COOH and *OCHO intermediates and enhance electron transfer while suppressing HER. Consequently, TL9/Cu-40% sustains >50% FE for CO₂RR even at −0.8 V, a potential at which HER typically dominates. In contrast, agglomerated Cu₂O in YEC/Cu-40% lowers the effective ECSA and exposes non-Cu sites that favor hydrogen evolution. These findings underscore that nanoscale dispersion and interfacial uniformity are decisive factors translating directly into CO₂RR activity and selectivity.

CV curves of TL9/Cu composites with varying Cu₂O loadings (10–50%) are presented in Fig. S11a. All samples show similar onset potentials near −0.8 V vs. RHE, suggesting that CO₂ activation thermodynamics are relatively unaffected by loading. However, the catalytic current increases markedly from 10 to 40%, reflecting enhanced ECSA (Fig. S11b). At 50%, the current slightly declines, implying that excessive Cu₂O induces nanoparticle agglomeration, which impedes charge transfer and site accessibility.⁶² The corresponding FEs (Fig. S11c–f) reveal that at 10% (Fig. S11c), H₂ is the dominant product, with limited CO and formate formation due to the low ECSA and exposure of uncoordinated N sites that favor HER.⁵⁸ Increasing the loading to 20% (Fig. S11d) enhances CO production, but formate selectivity remains low because Cu₂O agglomeration reduces the stabilized Cu⁺ fraction, favoring the *COOH pathway. At 30% (Fig. S11e), CO₂RR activity improves significantly, producing both CO and formate, attributed to a balanced Cu⁺/Cu⁰ ratio from more uniform Cu₂O coverage. The highest performance occurs at 40% (Fig. 10c), where optimal Cu⁺/Cu⁰ balance yields abundant active sites and an FE of ∼90% toward CO₂RR products. Further increasing the loading to 50% (Fig. S11f) lowers formate selectivity and increases H₂ (∼20%), consistent with agglomeration observed by SEM and H₂-TPR. This aggregation diminishes the stabilized Cu⁺ fraction, similar to the 20% sample, leading again to CO-dominated products.

Long-term electrolysis (Fig. 11) further demonstrates the stability advantage of TL9/Cu-40%. All long-term electrolysis performed under identical conditions at an applied potential of −0.6 V vs. RHE in CO₂-saturated 0.5 M KHCO₃ with a CO₂ flow rate of 20 sccm. The FE for TL9/Cu-40% remains above 60% after 24 h, with current density stabilizing at ∼1.5 mA cm⁻². This modest decline suggests partial surface restructuring yet retention of a significant fraction of active Cu⁺/Cu⁰ sites.⁶³ By contrast, YEC/Cu-40% exhibits rapid deactivation, with FE dropping from ∼25% to <15% and current density declining substantially, consistent with weaker Cu–support interactions leading to Cu₂O agglomeration and irreversible reduction to Cu⁰. Overall, these observations confirm that nitrogen defect anchoring in TL9 not only enhances CO₂RR activity and selectivity but also mitigates degradation, enabling sustained catalytic performance under prolonged operation.

Fig. 11 Stability test of (a) TL9/Cu-40% and (b) YEC/Cu-40% at 0.6 V vs. RHE using H-type cell in 0.5 M KHCO₃ electrolyte saturated with CO₂ (pH = 7.4) for 24 hours.

To further confirm the identity of the liquid-phase product, the post-electrolysis electrolyte was analyzed by ¹H NMR and HPLC. As shown in Fig. S13, a distinct resonance at ∼8.4 ppm, characteristic of formate, was observed in the ¹H NMR spectrum, with no detectable signals from other liquid products. Quantitative HPLC chromatogram of the post-electrolysis electrolyte (Fig. S14), together with the corresponding calibration curve (Fig. S15), further confirms formate as the dominant liquid product.

3.4 Post-CO₂RR surface functionalities

To directly verify the stabilizing effect of TL9 on Cu⁺ species, post-CO₂RR Cu LMM Auger XPS analysis was performed and compared with the pristine catalysts, as shown in Fig. S12. The relative fractions of Cu⁺, Cu⁰, and Cu²⁺ were quantified by peak deconvolution and are summarized in Table S6.

As shown in Fig. S12a, TL9/Cu-40% before CO₂RR consists predominantly of Cu⁺ species (∼64%), with only a minor contribution from Cu⁰ (∼14%). After 24 h of CO₂ electroreduction (Fig. S12a, bottom), the Cu LMM spectrum of TL9/Cu-40% shows that a substantial fraction of Cu⁺ is retained (∼36%), with only a moderate increase in the Cu⁰ component. Importantly, the Cu⁺ peak position and intensity remain clearly discernible, indicating that Cu⁺ is not fully reduced to metallic Cu during prolonged operation.

In contrast, YEC/Cu-40% (Fig. S12b) exhibits a markedly different behavior. Even before CO₂RR, YEC/Cu-40% contains a larger fraction of Cu⁰ (∼34%), consistent with weaker metal–support interactions. After CO₂RR, the Cu LMM spectrum shows a dominant Cu⁰ signal, with Cu⁺ becoming a minor component, evidencing extensive reduction of Cu⁺ to Cu⁰ during electrolysis.

Beyond demonstrating Cu⁺ stabilization, post-electrolysis Cu LMM Auger XPS analysis provides direct insight into the degradation mechanism. The observed decline in FE over extended electrolysis can be rationalized by several concurrent degradation processes. First, weakly stabilized Cu⁺ sites that are not strongly coordinated to nitrogen defects are preferentially reduced or lost during prolonged operation, whereas Cu⁺ species anchored through robust Cu–N_x interactions remain catalytically active.¹⁰ Second, sustained cathodic polarization induces electrochemical surface reconstruction, driven by local pH gradients, Cu atom migration, and dynamic restructuring of the catalyst surface.^64,65 Finally, these processes promote the gradual growth of metallic Cu domains, which tend to favor hydrogen evolution and CO formation over the formate pathway.⁶⁶ As a result, the relative contribution of C₁ products decreases with time, leading to the observed reduction in overall selectivity despite the persistence of strongly stabilized Cu–N_x active sites.

Therefore, the performance decay reflects bias-induced Cu⁺ evolution rather than failure of the Cu⁺ stabilization mechanism, and the superior long-term selectivity of TL9/Cu-40% compared to YEC/Cu-40% directly supports the role of nitrogen defects in mitigating electrochemical degradation.

4. Conclusion

In summary, N-doped carbon derived from tea leaves provides a defect-rich scaffold that anchors Cu₂O nanoparticles and stabilizes Cu⁺ species through strong Cu–N_x coordination. This interfacial stabilization enables a persistent Cu⁺/Cu⁰ synergy that drives efficient CO₂ electroreduction to CO and formate while mitigating hydrogen evolution and catalyst deactivation. The optimized TL9/Cu–40% catalyst exhibits uniform Cu₂O dispersion, high FE (∼90%), and long-term operational stability exceeding 24 h. In contrast, Cu supported on commercial activated carbon suffers from particle agglomeration and loss of Cu⁺ activity, underscoring the crucial role of metal–support interactions. By integrating surface defect engineering with sustainable biomass-derived carbons, this work redefines pathways toward scalable, selective, and stable CO₂ conversion. The demonstrated control over oxidation-state dynamics and dual product selectivity positions TL9/Cu as a model bifunctional system for the advanced electrocatalysts that merge material design with electrochemical functionality to unlock the next generation of CO₂-to-chemicals technologies.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI): Tables S1–S6, additional N₂ adsorption-desorption isotherms, TGA plots, XRD patterns of the composites with various Cu₂O loading on TL9, additional TEM and SEM images, H₂-TPR profiles, XPS survey spectrum, C 1s spectra, and O 1s spectra, correlation between Cu atomic percentage and Cu–Nₓ, additional catalytic performances, Cu LMM spectra before and after CO₂RR electrolysis, ¹H NMR and HPLC data. See DOI: https://doi.org/10.1039/d5ra08966b.

Acknowledgements

This research was supported by Thailand Science Research and Innovation (TSRI) Fundamental Fund, fiscal year 2025. P. A. acknowledges the graduate scholarship under the SIIT-JAIST Dual Doctoral Degree Program. P. S. acknowledges the support from Thammasat Post Doctoral Fellowship (Contract No. TUPD 4/2567).

References

1. R.-Z. Xiong, H.-M. Xu, H.-R. Zhu, Z.-J. Zhang and G.-R. Li, Recent progress in Cu-based electrocatalysts for CO₂ reduction, Chem. Eng. J., 2025, 505, 159210.
2. H. Xie, T. Wang, J. Liang, Q. Li and S. Sun, Cu-based nanocatalysts for electrochemical reduction of CO₂, Nano Today, 2018, 21, 41–54.
3. T.-C. Chou, C.-C. Chang, H.-L. Yu, W.-Y. Yu, C.-L. Dong and J.-J. Velasco-Vélez, et al., Controlling the Oxidation State of the Cu Electrode and Reaction Intermediates for Electrochemical CO₂ Reduction to Ethylene, J. Am. Chem. Soc., 2020, 142(6), 2857–2867.
4. H. Tabassum, X. Yang, R. Zou and G. Wu, Surface engineering of Cu catalysts for electrochemical reduction of CO₂ to value-added multi-carbon products, Chem Catal., 2022, 2(7), 1561–1593.
5. E. Schuler, M. Morana, P. A. Ermolich, K. Lüschen, A. J. Greer and S. F. R. Taylor, et al., Formate as a key intermediate in CO₂ utilization, Green Chem., 2022, 24(21), 8227–8258.
6. G. Centi, E. A. Quadrelli and S. Perathoner, Catalysis for CO₂ conversion: a key technology for rapid introduction of renewable energy in the value chain of chemical industries, Energy Environ. Sci., 2013, 6(6), 1711–1731.
7. M. Ronda-Lloret, G. Rothenberg and N. R. Shiju, A Critical Look at Direct Catalytic Hydrogenation of Carbon Dioxide to Olefins, ChemSusChem, 2019, 12(17), 3896–3914.
8. Y. Zhang, G. Jacobs, D. E. Sparks, M. E. Dry and B. H. Davis, CO and CO₂ hydrogenation study on supported cobalt Fischer–Tropsch synthesis catalysts, Catal. Today, 2002, 71(3–4), 411–418.
9. A. Vijayakumar, Y. Zhao, K. Wang, Y. Chao, H. Chen and C. Wang, et al., A Nitrogen-Doped Porous Carbon Supported Copper Catalyst from a Scalable One-Step Method for Efficient Carbon Dioxide Electroreduction, ChemElectroChem, 2023, 10(2), e202200817.
10. D. A. Bulushev, A. L. Chuvilin, V. I. Sobolev, S. G. Stolyarova, Y. V. Shubin and I. P. Asanov, et al., Copper on carbon materials: stabilization by nitrogen doping, J. Mater. Chem. A, 2017, 5(21), 10574–10583.
11. H. Ning, X. Wang, W. Wang, Q. Mao, Z. Yang and Q. Zhao, et al., Cubic Cu₂O on nitrogen-doped carbon shells for electrocatalytic CO₂ reduction to C₂H₄, Carbon, 2019, 146, 218–223.
12. D. Cheng, Z.-J. Zhao, G. Zhang, P. Yang, L. Li and H. Gao, et al., The nature of active sites for carbon dioxide electroreduction over oxide-derived copper catalysts, Nat. Commun., 2021, 12(1), 395.
13. K. Yue, Y. Qin, H. Huang, Z. Lv, M. Cai and Y. Su, et al., Stabilized Cu⁰-Cu¹⁺ dual sites in a cyanamide framework for selective CO₂ electroreduction to ethylene, Nat. Commun., 2024, 15(1), 7820.
14. D. Li, T. Liu, L. Huang, J. Wu, J. Li and L. Zhen, et al., Selective CO₂-to-formate electrochemical conversion with core–shell structured Cu₂O/Cu@C composites immobilized on nitrogen-doped graphene sheets, J. Mater. Chem. A, 2020, 8(35), 18302–18309.
15. S. S. Desa and K. Nueangnoraj, Enhanced Redox Activity of Anthraquinone on N-Doped Porous Carbon Electrode, ChemistrySelect, 2024, 9(21), e202303953.
16. T. Witoon, N. Kachaban, W. Donphai, P. Kidkhunthod, K. Faungnawakij and M. Chareonpanich, et al., Tuning of catalytic CO₂ hydrogenation by changing composition of CuO–ZnO–ZrO₂ catalysts, Energy Convers. Manage., 2016, 118, 21–31.
17. X. Zhang, X. Xia, I. Ivanov, X. Huang and B. E. Logan, Enhanced activated carbon cathode performance for microbial fuel cell by blending carbon black, Environ. Sci. Technol., 2014, 48(3), 2075–2081.
18. M. D. Donohue and G. L. Aranovich, Classification of Gibbs adsorption isotherms, Adv. Colloid Interface Sci., 1998, 76–77, 137–152.
19. V. Georgakilas, J. A. Perman, J. Tucek and R. Zboril, Broad Family of Carbon Nanoallotropes: Classification, Chemistry, and Applications of Fullerenes, Carbon Dots, Nanotubes, Graphene, Nanodiamonds, and Combined Superstructures, Chem. Rev., 2015, 115(11), 4744–4822.
20. V. Harnchana, S. Chaiyachad, S. Pimanpang, C. Saiyasombat, P. Srepusharawoot and V. Amornkitbamrung, et al., Hierarchical Fe₃O₄-reduced graphene oxide nanocomposite grown on NaCl crystals for triiodide reduction in dye-sensitized solar cells, Sci. Rep., 2019, 9(1), 1494.
21. A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner and U. Pöschl, Raman microspectroscopy of soot and related carbonaceous materials: spectral analysis and structural information, Carbon, 2005, 43(8), 1731–1742.
22. M. A. S. M. Haniff, S. M. Hafiz, K. A. Wahid, Z. Endut, M. I. Syono and N. M. Huang, et al., Nitrogen-doped multiwalled carbon nanotubes decorated with copper(I) oxide nanoparticles with enhanced capacitive properties, J. Mater. Sci., 2017, 52(11), 6280–6290.
23. M. W. Iqbal, A. K. Singh, M. Z. Iqbal and J. Eom, Raman fingerprint of doping due to metal adsorbates on graphene, J. Phys.: Condens. Matter, 2012, 24(33), 335301.
24. F. Farivar, P. L. Yap, R. U. Karunagaran, D. Losic, F. Farivar and P. Lay Yap, et al., Thermogravimetric Analysis (TGA) of Graphene Materials: Effect of Particle Size of Graphene, Graphene Oxide and Graphite on Thermal Parameters, C, 2021, 7(2), 41.
25. M. Suominen and T. Kallio, What We Currently Know about Carbon-Supported Metal and Metal Oxide Nanomaterials in Electrochemical CO₂ Reduction, ChemElectroChem, 2021, 8(13), 2397–2406.
26. S. V. Vassilev, D. Baxter, L. K. Andersen and C. G. Vassileva, An overview of the composition and application of biomass ash. Part 1. Phase–mineral and chemical composition and classification, Fuel, 2013, 105, 40–76.
27. J. Zhai, I. T. Burke, W. M. Mayes and D. I. Stewart, New insights into biomass combustion ash categorisation: a phylogenetic analysis, Fuel, 2021, 287, 119469.
28. X. Zhu, Y. Zeng, X. Zhao, D. Liu, W. Lei and S. Lu, Biomass-Derived Carbon and Their Composites for Supercapacitor Applications: Sources, Functions, and Mechanisms, EcoEnergy, 2025, 3(3), e70000.
29. L. Vivas, I. Chi-Duran, J. Enríquez, N. Barraza, D. P. Singh and L. Vivas, et al., Ascorbic acid based controlled growth of various Cu and Cu₂O nanostructures, Mater. Res. Express, 2019, 6(6), 065033.
30. Y. Shen, L. Qian, Q. Xu, S. Wang, Y. Chen and H. Lu, et al., N-doped Cu₂O with the tunable Cu⁰ and Cu⁺ sites for selective CO₂ electrochemical reduction to ethylene, J. Environ. Sci., 2025, 150, 246–253.
31. B. Ahn, L. Bosetti and M. Mazzotti, Secondary Nucleation by Interparticle Energies. II. Kinetics, Cryst. Growth Des., 2021, 22(1), 74–86.
32. N. T. K. Thanh, N. Maclean and S. Mahiddine, Mechanisms of Nucleation and Growth of Nanoparticles in Solution, Chem. Rev., 2014, 114(15), 7610–7630.
33. R. Poreddy, C. Engelbrekt and A. Riisager, Copper oxide as efficient catalyst for oxidative dehydrogenation of alcohols with air, Catal. Sci. Technol., 2015, 5(4), 2467–2477.
34. F. Qi, J. Peng, Z. Liang, J. Guo, J. Liu and T. Fang, et al., Strong metal-support interaction (SMSI) in environmental catalysis: mechanisms, application, regulation strategies, and breakthroughs, Environ. Sci. Ecotechnology, 2024, 22, 100443.
35. H. Silva, C. Mateos-Pedrero, P. Ribeirinha, M. Boaventura, A. Mendes and H. Silva, et al., Low-temperature methanol steam reforming kinetics over a novel CuZrDyAl catalyst, React. Kinet., Mech. Catal., 2015, 115(1), 321–339.
36. Z. Zhang, H. Che, J. Gao, Y. Wang, X. She and J. Sun, et al., Shape-controlled synthesis of Cu₂O microparticles and their catalytic performances in the Rochow reaction, Catal. Sci. Technol., 2012, 2(6), 1207–1212.
37. Y.-H. Lee, Y.-F. Lee, K.-H. Chang and C.-C. Hu, Synthesis of N-doped carbon nanosheets from collagen for electrochemical energy storage/conversion systems, Electrochem. Commun., 2011, 13(1), 50–53.
38. M. Vujković, N. Gavrilov, I. Pašti, J. Krstić, J. Travas-Sejdic and G. Ćirić-Marjanović, et al., Superior capacitive and electrocatalytic properties of carbonized nanostructured polyaniline upon a low-temperature hydrothermal treatment, Carbon, 2013, 64, 472–486.
39. D. X. Oh, S. Shin, C. Lim and D. S. Hwang, Dopamine-Mediated Sclerotization of Regenerated Chitin in Ionic Liquid, Materials, 2013, 6(9), 3826–3839.
40. H. Jiang, L. Liu, K. Zhao, Z. Liu, X. Zhang and S. Hu, Effect of pyridinic- and pyrrolic-nitrogen on electrochemical performance of Pd for formic acid electrooxidation, Electrochim. Acta, 2020, 337, 135758.
41. G. Greczynski, D. Primetzhofer, J. Lu and L. Hultman, Core-level spectra and binding energies of transition metal nitrides by non-destructive X-ray photoelectron spectroscopy through capping layers, Appl. Surf. Sci., 2017, 396, 347–358.
42. Y.-M. Chang, J. Leu, B.-H. Lin, Y.-L. Wang and Y.-L. Cheng, Comparison of H₂ and NH₃ Treatments for Copper Interconnects, Adv. Mater. Sci. Eng., 2013, 2013(1), 825195.
43. L. Zhu, X. Cao, J. Xiao, S. Ma and S. Ta, Structure and photocatalytic properties of TiO₂/Cu₃N composite films prepared by magnetron sputtering, Mater. Today Commun., 2021, 26, 101739.
44. H. Perez, V. Jorda, P. Bonville, J. Vigneron, M. Frégnaux and A. Etcheberry, et al., Synthesis and Characterization of Carbon/Nitrogen/Iron Based Nanoparticles by Laser Pyrolysis as Non-Noble Metal Electrocatalysts for Oxygen Reduction, C, 2018, 4(3), 43.
45. D. Divya, M. Nandhagopal and S. Thennarasu, Deprotonation of the –NH proton from a pyrrole moiety enables concentration-dependent colorimetric and fluorescence quenching of silver(I) ions, New J. Chem., 2025, 49(10), 3918–3929.
46. Q. Lu, V. McKee and J. Nelson, Redox-linked pyrrole NH deprotonation in a dicopper azacryptate system, J. Chem. Soc. Chem. Commun., 1994,(5), 649–651.
47. D. Hulicova-Jurcakova, M. Seredych, G. Q. Lu and T. J. Bandosz, Combined Effect of Nitrogen- and Oxygen-Containing Functional Groups of Microporous Activated Carbon on its Electrochemical Performance in Supercapacitors, Adv. Funct. Mater., 2009, 19(3), 438–447.
48. M. F. R. Pereira, S. F. Soares, J. J. M. Órfão and J. L. Figueiredo, Adsorption of dyes on activated carbons: influence of surface chemical groups, Carbon, 2003, 41(4), 811–821.
49. R. Xie, Y. Jin, Y. Chen and W. Jiang, The importance of surface functional groups in the adsorption of copper onto walnut shell derived activated carbon, Water Sci. Technol., 2017, 76(11–12), 3022–3034.
50. J.-Y. Huang, X.-D. Zhang, H. Yang, Q.-W. Liu, W.-W. Yuan and W.-C. Lai, et al., Pulsed Strategy Steers the Structural Evolution of Cu Metal–Organic Framework for CO₂ Reduction to Methane, Chem.–Eur. J., 2025, 31(35), e202500744.
51. L. Xu, Y. Xiao, Z.-X. Yu, Y. Yang, C. Yan and J.-Q. Huang, Revisiting the Electrochemical Impedance Spectroscopy of Porous Electrodes in Li-ion Batteries by Employing Reference Electrode, Angew. Chem., Int. Ed., 2024, 63(41), e202406054.
52. L. Shreenivasa, K. Yogesh, S. A. Prashanth, R. Viswanatha and S. Ashoka, Enhancement of cycling stability and capacity of lithium secondary battery by engineering highly porous AlV₃O₉, J. Mater. Sci., 2019, 55(4), 1648–1658.
53. A. Lasia, Impedance of Porous Electrodes, in Electrochemical Impedance Spectroscopy and its Applications, Springer, New York, 2014, pp. 203–250.
54. Y. Zhao, D. Su, W. Dong, X. Xu, X. Zhang and Y. Hu, High crystallinity Sn crystals on Ni foam: an ideal bimetallic catalyst for the electroreduction of carbon dioxide to syngas, RSC Adv., 2020, 10(64), 39026–39032.
55. C. Guo, S. Liu, Z. Chen, B. Li, L. Chen and C. V. Singh, et al., How does mass transfer influence electrochemical carbon dioxide reduction reaction? A case study of Ni molecular catalyst supported on carbon, Chem. Commun., 2021, 57(11), 1384–1387.
56. J. Dong, Y. Liu, J. Pei, H. Li, S. Ji and L. Shi, et al., Continuous electroproduction of formate via CO₂ reduction on local symmetry-broken single-atom catalysts, Nat. Commun., 2023, 14(1), 6849.
57. X. Wang, S. Ding, X. Feng and Y. Zhu, High stability copper clusters anchored on N-doped carbon nanosheets for efficient CO₂ electroreduction to HCOOH, J. Colloid Interface Sci., 2024, 653, 741–748.
58. G.-f. Long, K. Wan, M.-y. Liu, Z.-x Liang, J.-h. Piao and P. Tsiakaras, Active sites and mechanism on nitrogen-doped carbon catalyst for hydrogen evolution reaction, J. Catal., 2017, 348, 151–159.
59. T. Sasipatworakarn, D. Pikulrat, K. Homlamai, S. Duangdangchote and M. Sawangphruk, Selective conversion of carbon dioxide to formate using few-layer nitrogen-doped graphene on copper foam with enhanced suppression of the hydrogen evolution reaction, Sustainable Energy Fuels, 2024, 8(14), 3164–3173.
60. C. Jiménez, M. I. Cerrillo, F. Martínez, R. Camarillo and J. Rincón, Effect of carbon support on the catalytic activity of copper-based catalyst in CO₂ electroreduction, Sep. Purif. Technol., 2020, 248, 117083.
61. O. A. Baturina, Q. Lu, M. A. Padilla, L. Xin, W. Li and A. Serov, et al., CO₂ Electroreduction to Hydrocarbons on Carbon-Supported Cu Nanoparticles, ACS Catal., 2014, 4(10), 3682–3695.
62. X. Tuaev, X. Tuaev, S. Rudi, S. Rudi, P. Strasser and P. Strasser, The impact of the morphology of the carbon support on the activity and stability of nanoparticle fuel cell catalysts, Catal. Sci. Technol., 2016, 6(23), 8276–8288.
63. Z. Lian, F. Dattila, N. López, Z. Lian, F. Dattila and N. López, Stability and lifetime of diffusion-trapped oxygen in oxide-derived copper CO₂ reduction electrocatalysts, Nat. Catal., 2024, 7(4), 401–411.
64. Z. Han, D. Han, Z. Chen, J. Gao, G. Jiang and X. Wang, et al., Steering surface reconstruction of copper with electrolyte additives for CO₂ electroreduction, Nat. Commun., 2022, 13(1), 3158.
65. Y. Zhong, Z. Sun, B. Y. Xia and Y. Su, Structural Reconstruction of Copper-Based Catalysts in CO₂ Electroreduction Reaction: A Comprehensive Review, Chem.–Eur. J., 2025, 31(27), e202500770.
66. S. Nitopi, E. Bertheussen, S. B. Scott, X. Liu, A. K. Engstfeld and S. Horch, et al., Progress and Perspectives of Electrochemical CO₂ Reduction on Copper in Aqueous Electrolyte, Chem. Rev., 2019, 119(12), 7610–7672.

---