Due to human intervention in the nitrogen cycle, including the widespread use of nitrogen-containing fertilizers and the discharge of domestic or industrial wastewater, nitrate pollution has affected both surface and groundwater [1,2]. Apart from harming aquatic ecosystems, nitrate pollution in water can lead to severe human health issues [3]. In response to this, the World Health Organization (WHO) has set the maximum permissible nitrate concentration in drinking water at 10 mg/L [4]. Some technologies such as biological denitrification, ion exchange, reverse osmosis, and electrodialysis have been developed to remediate nitrate-rich wastewater [5-8]. Nevertheless, they come with inevitable drawbacks, including secondary pollution, slow reaction rates, and high reaction costs, making them unsuitable for large-scale applications [9,10]. In recent years, electrocatalytic nitrate reduction reaction (NO₃⁻RR), utilizing electrons as a clean reducing agent to convert nitrate into ammonia or harmless nitrogen gas, has emerged as a promising method for mitigating nitrate pollution [11,12]. However, considering the fact that many practical sources of nitrate wastewater, such as natural/industrial wastewater, have low concentrations ranging from hundreds to thousands of ppm [13,14]. It is crucial for the development of advanced electrocatalysts to enhance nitrate reduction at low initial NO₃⁻-N concentrations.

To date, various transition metal catalysts, such as Cu, Fe, Ni, and Co, have been studied for their application in the electrochemical reduction of NO₃⁻-N in water [4,15-18]. Among them, low-cost and non-toxic copper-based catalysts exhibit strong NO₃⁻-N adsorption capability, especially in the dilute nitrate system, which demonstrates the highest reaction kinetics in the rate-determining step in NO₃⁻RR (conversion of NO₃⁻-N to NO₂⁻-N) [19]. However, single-component copper-based catalysts suffer from poor catalyst stability and low ammonia selectivity due to competitive adsorption of intermediate products. To address these issues, researchers have proposed various solutions, such as doping, alloy strategies, and the preparation of copper-based oxides, to enhance nitrate reduction [20-22]. Recently, Wang et al. [18] discovered that copper-based oxides undergo in-situ transformation to Cu⁰ during electrochemical reactions, and their performance surpasses that of CuO and Cu₂O. Ren et al. [23] developed a multi-phase heterostructure material Cu/Pd/CuO_x through in-situ restructuring, exhibiting excellent NO₃⁻RR catalytic activity. Under conditions of −1.3 V vs. SCE, ammonia production rate and ammonia Faradaic efficiency reached 1510.3 µg h⁻¹ mg_cat.⁻¹ and 84.04%, respectively. In these catalysts, the in-situ reconstructed Cu⁰ serves as the catalytically active site for NO₃⁻RR. Furthermore, the electrochemical reduction potential can drive the generation of defects, leading to significant structural disorder [24,25]. However, the impact of such structural disorder has been seldom explored on NO₃⁻RR. The introduction of substantial structural perturbations induces lattice distortion, resulting in the amorphization of materials. Amorphous materials with disordered structures have been widely applied in many electrocatalytic reactions due to their high structural flexibility and exposure of abundant unsaturated active sites [26,27]. Nonetheless, as indicated in previously reported works, structural amorphization often involves complex processing. It typically requires hazardous gas treatment or high-energy ion implantation at elevated temperatures [28,29]. Consequently, constructing a catalyst with a disordered structure through a simple approach for exploring the electrocatalytic nitrate reduction performance at low NO₃⁻-N concentrations is of vital importance.

In this study, a Cu₃N NRs pre-catalyst with a nanorod structure was synthesized through chemical oxidation and nitrification. Under electrocatalytic nitrate reduction conditions, it underwent in-situ reconstruction to form a Cu-based catalyst with a disordered structure of Cu⁰ (re-Cu NRs). Structural characterization and electrochemical analysis revealed that the reconstructed Cu⁰, acting as the active catalytic species for the conversion of NO₃⁻-N to NH₄⁺-N, exhibited amorphous boundaries that exposed numerous unsaturated active sites. As anticipated, the reconstructed re-Cu NRs material demonstrated excellent catalytic performance of NO₃⁻RR and long-term stability at low concentrations of NO₃⁻-N. The use of ¹⁵N labeling confirmed the generation of NH₄⁺ during NO₃⁻RR. The detection of intermediate products through online differential electrochemical mass spectrometry (DEMS) was employed to infer possible reaction pathways. Density functional theory (DFT) calculations confirmed that the amorphous structure of the re-Cu NRs catalyst facilitated nitrate adsorption and suppression of co-existing hydrogen evolution reactions. Finally, this work combined electrocatalytic NO₃⁻RR and electrochlorination through the conversion path of NO₃⁻-N → NH₄⁺-N → N₂, achieving efficient and environmentally benign nitrate treatment.

The synthesis strategy for re-Cu NRs is illustrated in Fig. S1 (Supporting information). Initially, copper foam underwent wet chemical oxidation in a mixed solution of NaOH and (NH₄)₂S₂O₈, resulting in vertically aligned Cu(OH)₂ nanorods (NRs) arrays. Subsequently, nitrification was performed using urea in a nitrogen atmosphere at 350 ℃ to form Cu₃N NRs. The prepared Cu₃N NRs was then employed as a pre-catalyst directly for electrochemical nitrate reduction, leading to the in-situ reconstruction of re-Cu NRs under electrocatalytic conditions. Morphological characterization of the prepared material was conducted using scanning electron microscopy (SEM). Fig. 1 depicts the surface morphology of Cu(OH)₂NRs/CF, Cu₃N NRs, and re-Cu NRs at various magnifications. As shown in Figs. 1a-c, after chemical oxidation, Cu(OH)₂ NRs with a length of approximately 5–8 µm and a diameter of about 250 nm uniformly grew on the CF surface. Cu₃N NRs/CF maintained the original morphology after nitrification at 350 ℃ for 120 min, revealing that each nanorod was composed of stacked nanoparticles (Figs. 1d-f). Figs. 1g-i display the further changes in re-Cu NRs after electrochemical reduction. The reconstructed nanorod surface became rougher, exposing a larger specific surface area, which facilitates enhanced charge transfer and is advantageous for electrocatalytic applications [19].

Further morphological and structural analysis of re-Cu NRs was conducted using transmission electron microscopy (TEM). Fig. 2a illustrates the overall structure of re-Cu NRs, revealing that the nanorods were composed of numerous nanoparticles with sizes around 25-30 nm. High-resolution transmission electron microscopy (HRTEM) provides clear insight into the ordered crystal structure inside the nanoparticles and the amorphous structure at the interface (marked by the white dashed curve in Fig. 2b. Simultaneously, a clear lattice spacing of 0.210 nm corresponding to the crystal plane (111) of Cu can be observed in the crystalline region. For Cu₃N NRs, the crystal structure of Cu₃N is easily identified with a lattice spacing of 0.382 nm corresponding to the crystal plane (100) of Cu₃N (Fig. S2 in Supporting information) [30]. The transition from the crystal plane (100) of Cu₃N to the crystal plane (111) of Cu confirms the in-situ reconstruction of Cu₃N NRs during the electrochemical reduction process. Moreover, it is worth noting that the in-situ reconstruction resulted in the amorphous boundary region around the nanoparticles, contributing to enhanced electrocatalytic activity. Compared to crystal structures, the disordered atomic arrangement in amorphous structures exposes abundant active sites with unsaturated structures, favouring the adsorption and conversion of NO₃⁻-N [31,32]. Subsequent scanning transmission electron microscopy-energy dispersive spectroscopy (STEM-EDS) results (Fig. S3 in Supporting information) also show a decrease in the nitrogen content on the surface of Cu₃N NRs electrodes, validating the in-situ transformation from Cu₃N to Cu during the electrochemical reduction process.

To investigate the chemical changes during the in-situ reduction of Cu₃N NRAs, X-ray diffraction (XRD) and high-resolution X-ray photoelectron spectroscopy (XPS) tests were conducted. Fig. 2c displays the XRD patterns of Cu₃N NRs and re-Cu NRs. The peaks at 23.3°, 33.2°, 40.9°, 47.6°, 53.6°, and 69.6° corresponded well to the planes (100), (110), (111), (200), (210), and (220) of the cubic structure of Cu₃N (JCPDFS No. 47–1088), consistent with previous reports [30,33]. After electrochemical reduction, the peaks of Cu₃N disappeared, leaving three characteristic peaks at 43.29°, 50.43°, and 74.13°, corresponding to the planes (111), (200), and (220) of the cubic Cu (JCPDS No. 04–0836). This is in accordance with the TEM analysis results.

To further test the surface composition and chemical states of these two samples, XPS analysis was performed. As shown in Fig. 2d, the two peaks cantered at 932.2 eV and 952.0 eV in the Cu 2p spectrum could be assigned to Cu 2p_3/2 and Cu 2p_1/2, respectively. To further distinguish between Cu⁺ and Cu⁰, the chemical state of copper was identified with auger electron spectroscopy (AES) (Fig. 2e). There is only one characteristic peak centered on 917.1 eV from Cu⁺ in the Cu₃N NRs [34,35]. After electrochemical reduction, a new peak cantered at 918.8 eV appeared on the surface of re-Cu NRs, corresponding to Cu⁰, indicating the conversion of Cu⁺ to Cu⁰ occurring during the reconstruction process [18,23]. It is noteworthy that the Cu 2p XPS spectrum of re-Cu NRs exhibited a negative shift compared to Cu₃N NRs (Fig. 2d). In contrast, the Cu⁰ Auger peak in the re-Cu NRs electrode had a positive shift of 0.2 eV compared to pure Cu metal (918.6 eV), indicating the presence of electron transfer at the re-Cu NRs electrode interface [19]. Simultaneously, in the N 1s spectrum (Fig. 2f), the peak corresponding to the Cu-N bond at a binding energy of 397.3 eV disappeared [30,36]. The above results indicate that the re-Cu NRs with amorphous boundaries are the active substances during electrocatalytic nitrate reduction.

In a single-chamber electrolytic cell, the electrocatalytic performance of the re-Cu NRs electrode was evaluated using Ag/AgCl and Pt foil as reference electrode and counter electrode, respectively. Linear sweep voltammetry (LSV) polarization curves for different electrodes in a 50 mg/L NO₃⁻-N and 0.05 mol/L Na₂SO₄ solution are shown in Fig. 3a. Compared to CF substrate and Cu(OH)₂ NRs/CF electrode, the re-Cu NRs electrode exhibited higher current density, indicating superior electrocatalytic nitrate reduction performance. The double-layer capacitance (C_dl) values of re-Cu NRs were calculated from the cyclic voltammetry (CV) curves at different scan rates (1–5 mV/s) (Fig. S4 in Supporting information). As shown in Fig. 3b, the C_dl value of re-Cu NRs was 123.12 mF/cm², which was 1.54 times that of Cu(OH)₂ NRs/CF (79.9 mF/cm²), and 19.05 times that of CF substrate (6.46 mF/cm²). The larger electrochemical active surface area of re-Cu NRs may be attributed to the amorphous boundaries formed by structural disorder, exposing more abundant active sites [37]. Furthermore, electrochemical impedance spectroscopy (EIS) measurements were conducted to study the electron transfer rate of the prepared re-Cu NRs electrode (Fig. 3c). The Nyquist plot for re-Cu NRs exhibited a smaller arc diameter compared to Cu(OH)₂ NRs/CF electrode and CF substrate, demonstrating that the charge transfer resistance of re-Cu NRs was minimal. This suggests fast electron transfer rates during the electrochemical reaction, contributing to superior reaction kinetics and enhanced catalytic activity for NO₃⁻RR [38,39].

To further explore the electrocatalytic performance of re-Cu NRs, electrocatalytic NO₃⁻RR experiments were conducted at different applied potentials ranging from −0.9 V to −1.4 V. The concentrations of reactants and reduction products (NO₃⁻-N, NO₂⁻-N, and NH₄⁺-N) were detected using spectrophotometry, and corresponding concentration-absorbance calibration curves were plotted (Fig. S5 in Supporting information). As shown in Fig. 3d, with the increase in applied potential, the conversion rate of NO₃⁻-N continually increased, reaching 100% at −1.2 V and gradually stabilizing. Additionally, the electrocatalytic nitrate reduction process at different applied potentials conformed to the pseudo-first-order kinetic model (Fig. S6 in Supporting information). With the increase in working potential, the reduction rate of NO₃⁻-N and the corresponding pseudo-first-order kinetic constant (k) gradually increased (from 0.01063 min⁻¹ to 0.05231 min⁻¹). As the applied potential increased from −0.9 V to −1.4 V, the NH₄⁺-N selectivity gradually increased, reaching 99.60% at −1.2 V, while the NO₂⁻-N selectivity gradually decreased (Fig. S7 in Supporting information). As seen in Fig. 3e, the yield and Faradaic efficiency of NH₃ at different applied potentials exhibited a volcano-shaped curve, reaching peak values of 238.87 µg h⁻¹ cm⁻² and 44.30% at −1.2 V, respectively. This might be attributed to the intensified HER with the increase in applied potential, leading to a decrease in Faradaic efficiency [14]. Therefore, −1.2 V was identified as the most suitable applied potential for the electrocatalytic reduction of nitrate by re-Cu NRs.

Considering the practical application of re-Cu NRs in real wastewater, the impact of different initial concentrations of nitrate on electrocatalytic NO₃⁻RR was investigated. As shown in Fig. 3f, with the initial nitrate concentration increasing from 25 mg/L to 150 mg/L, the conversion rate of NO₃⁻-N decreased from 100% to 72.8%. This is possibly due to the limited active sites on the electrode surface, where excess NO₃⁻-N might not fully contact the electrode, restricting its adsorption on the active sites [40]. In addition, with the increase in initial NO₃⁻-N concentration, the selectivity of NH₄⁺-N gradually decreased, and the yield of NH₃ decreased from 238.87 µg h⁻¹ cm⁻² and stabilized at 165.23 µg h⁻¹ cm⁻². This phenomenon suggests that excessively high concentrations of NO₃⁻-N are unfavourable for NH₄⁺-N production. This could be attributed to the competition between NO₃⁻-N and NO₂⁻-N for reaction centres, where some NO₂⁻-N was displaced from the reaction centres before being hydrogenated to form NH₄⁺-N [41]. The re-Cu NRs exhibited higher nitrate conversion rates and ammonia yields at low concentrations, which could be directly applied to the in-situ remediation of water bodies with low-concentration nitrate pollution.

To explore the source of ammonia in electrochemical NO₃⁻RR and eliminate the interference of the external environment on the experimental results, several comparative experiments were carried out. In the comparative experiments conducted in a solution containing only Na₂SO₄, no ammonia generation was detected, proving that ammonia did not originate from the electrocatalyst (Fig. 3g). To further identify the source of ammonia, the isotope labelling experiment was conducted. As shown in the ¹H NMR spectrum (Fig. 3h) using K¹⁴NO₃ and K¹⁵NO₃ as reactants for the electrolytic reaction, the detected ¹⁴NH₄⁺−¹⁴N exhibited a typical triple-peak pattern, while ¹⁵NH₄⁺−¹⁵N showed a characteristic double-peak pattern [16,23]. This result affirmed that the generation of NH₄⁺ indeed originated from electrochemical NO₃⁻RR. Additionally, maleic acid (C₄H₄O₄) was used as an internal standard to quantify the generated NH₄⁺-N based on the standard curve of the integration area versus NH₄⁺ concentration (NH₄⁺-N/C₄H₄O₄) (Fig. S8 in Supporting information). The results revealed that the concentrations of ¹⁵NH₄⁺−¹⁵N (48.89 mg/L) and ¹⁴NH₄⁺−¹⁴N (49.02 mg/L) produced after 120 min of electrocatalytic nitrate reduction were very close to the determined concentrations by UV–vis spectrophotometry, confirming the accuracy of this method.

To elucidate the reaction pathways of the electrocatalytic process, online DEMS was employed to detect intermediates and products generated at the re-Cu NRs cathode during the electrocatalytic NO₃⁻RR process. Fig. 3i displays the m/z signals detected by online DEMS at an applied potential of −1.2 V, corresponding to species NO, NH, NH₂, and NH₃ with m/z values of 30, 15, 16, and 17, respectively. Moreover, no signal of NHOH (m/z = 32) was observed in the DEMS detection results, allowing us to propose the NO₃⁻RR reaction pathway on the surface of re-Cu NRs and used for theoretical calculations.

DFT calculations have been further conducted to demonstrate the significant contribution of the amorphous structure of the re-Cu NRs catalysts to the NO₃⁻RR. The schematic diagram of the process from Cu₃N NRs to re-Cu NRs is shown in Fig. 4a. In this process, the Cu₃N precatalyst suffers from the N leaching phenomenon, in which the N atoms are detached from the Cu₃N structure into the solution to form NH₄⁺-N (Fig. S9 in Supporting information), and finally, the Cu₃N NRs is reconfigured as re-Cu NRs with an amorphous structure. A computational model of re-Cu NRs with an amorphous structure was simulated with DFT, and Cu (111) was selected as the theoretical computational model of the Cu samples for comparison, considering the lattice spacing of the Cu (111) facet shown in the TEM. The primary competitive reaction for the electrocatalytic nitrate reduction reaction was HER, so the energies of the hydrogen evolution reactions were determined. As shown in Fig. 4b, the Gibbs free energy for the formation of H₂ on re-Cu NRs was 0.75 eV, which was above the energy required for Cu (111) (0.59 eV), indicating that re-Cu NRs had an inhibition effect on H₂ generation [42]. Moreover, based on the DEMS derived NO₃⁻RR reaction pathway on the surface of re-Cu NRs, the free energies of all the reduction intermediates on the surface of re-Cu NRs and Cu (111) were calculated (Fig. 4c). The model of the most stable adsorption of all intermediates on re-Cu NRs and Cu (111) surfaces is shown in Fig. S10 (Supporting information). During the process, NO₃⁻ was initially adsorbed on the catalyst surface, forming adsorbed nitrate (*NO₃⁻). As can be seen in Fig. 4c, the free energy of NO₃⁻ to form *NO₃⁻ on the Cu (111) surface was 0.52 eV, much higher than that of the re-Cu NRs (−1.70 eV), suggesting that the amorphous structure could promote the adsorption of NO₃⁻ on the surface of the catalyst and accelerate the subsequent reaction. Then, the N—O bond is continuously cleaved by proton-coupled electron transfer to form *NO₂ and *NO. Thereafter, the *NO intermediates were gradually converted to *N, *NH, *NH₂, and finally *NH₃ under hydrogenation [18]. It is noteworthy that the desorption of NH₃ from the catalyst surface is a potential determining step in the overall nitrate reduction process. The energy barrier of 0.61 eV is required at the re-Cu NRs for the formation of NH₃, lower than that at the Cu (111) surface (0.84 eV), indicating that the desorption of *NH₃ more easily proceeds at the surface of re-Cu NRs with an amorphous structure [43]. Therefore, DFT calculations of NO₃⁻RR and HER simulations showed that re-Cu NRs with an amorphous boundary not only enhanced NO₃⁻ adsorption but also optimized the adsorption energy of the intermediates (*NH₃ → NH₃) and inhibited the formation of H₂, leading to enhanced intrinsic electrocatalytic NO₃⁻RR activity.

To achieve complete conversion of NO₃⁻-N to harmless N₂, electrochemical NO₃⁻RR experiments were conducted in the presence of Cl⁻. Breakpoint chlorination is a widely used method for drinking water treatment, where the anodic in-situ oxidation of Cl⁻ generates active chlorine species (HOCl/ClO⁻), facilitating the conversion of NH₄⁺-N to N₂ [44,45]. Therefore, we selectively investigated the effect of NaCl concentration on NO₃⁻-N conversion, NO₂⁻-N generation, NH₄⁺-N generation and N₂ generation in the three-electrode device with Ag/AgCl and Pt foil as the reference and counter electrodes, respectively, and re-Cu NRs as working electrode. As shown in Fig. 5a, in the electrolyte without NaCl, the removal efficiency of NO₃⁻-N reached 100%, and it was not affected by the increase in NaCl concentration. As can be seen in Fig. 5b, the NO₂⁻-N production reached a maximum value of 20.09 mg/L with 20 min in the absence of NaCl, it decreased over time and eventually failed to be detectable. Meanwhile, the selectivity of NH₄⁺-N continuously increased, reaching 98.94% at 120 min. The product distribution of the re-Cu NRs electrode in NaCl electrolyte during the electrocatalytic nitrate reduction process was further analysed to elucidate the influence of NaCl (Figs. 5c-e). With the rise in NaCl concentration, the generation of NH₄⁺-N and NO₂⁻-N was significantly suppressed. When the NaCl concentration was 0.02, 0.05, and 0.08 mol/L, the selectivity for N₂ was 51.66%, 74.88%, and 99.41%, respectively. After electrolysis in a solution containing 0.08 mol/L NaCl for 120 min, no residual NO₂⁻-N was detected. It is noteworthy that further increasing the sodium chloride concentration to 0.10 mol/L did not further improve the selectivity for N₂ (Fig. 5f). Therefore, re-Cu NRs electrodes combined with anodic chlorine oxidation could effectively remove NO₃⁻-N and NO₂⁻-N and simultaneously convert NH₄⁺-N to harmless N₂ in an electrolyte containing 0.08 mol/L NaCl.

Table S1 (Supportting information) and Fig. 6a summarize and compare the removal efficiency for NO₃⁻-N and N₂ selectivity of Cu-based catalysts as reported in recent studies. Through the designed pathway combined with electrochlorination, the re-Cu NRs cathode achieved higher NO₃⁻-N removal efficiency and N₂ selectivity in a short time compared with other catalytic materials. The enhanced electrocatalytic NO₃⁻RR activity of re-Cu NRs could be mainly attributed to the in-situ reconstruction of Cu₃N NRs. During the in-situ electrochemical reduction process, Cu₃N NRs could transform into re-Cu NRs with amorphous boundaries, where Cu⁰ species acted as active sites to enhance NO₃⁻RR performance. Furthermore, in-situ reconstruction resulted in the formation of amorphous boundaries around nanoparticles with more defects and catalytic sites, thereby facilitating the adsorption, dissociation, and conversion of electrocatalytic intermediates [26,31,32].

Transition metal electrodes are prone to corrosion during electrolysis, leading to leaching of metal ions and secondary metal contamination. This is the reason why the stability of electrocatalysts is crucial in practical applications. To assess the stability of re-Cu NRs, cyclic experiments were conducted under optimal conditions. As shown in Fig. 6b, during 30 cycles of electrocatalytic nitrate reduction in Na₂SO₄—NaCl electrolyte, the re-Cu NRs cathode exhibited high catalytic stability, maintaining a conversion rate of nitrate between 90%−100%, and an average N₂ selectivity above 90%. The inductively coupled plasma mass spectrometry (ICP-MS) results demonstrated a low leaching of copper ions in the electrolyte after the reaction (Fig. S11 in Supporting information). Moreover, the Cu 2p profiles of fresh and used re-Cu NRs catalysts exhibited Cu 2p_3/2 and Cu 2p_1/2 peaks at 932.2 eV and 952.0 eV, respectively, and the Cu⁰ peak was still clearly visualized in the Cu LMM profile (Fig. 6c and its inset). It indicates that the Cu⁰ formation by in-situ reconfiguration could exist stably during electrocatalytic reduction, which is verified to have good stability. TEM characterization of the re-Cu NRs after the electrocatalytic reaction revealed that the re-Cu NRs still maintained its original nanorod morphology, demonstrating that the structure of the material has not changed after cycling (Fig. 6d). These findings indicate that the re-Cu NRs catalyst, with amorphous boundaries formed after reconstruction, exhibited excellent stability in multiple cycles.

In summary, a re-Cu NRs catalyst with an amorphous structure was prepared through in-situ electrochemical reconstruction. The well-defined three-dimensional nanorods structure and the amorphous boundaries on the surface expose abundant active sites, improving the performance of adsorption and conversion of NO₃⁻-N at low concentrations. The results of electrocatalytic NO₃⁻RR indicated that the re-Cu NRs electrode reached 100% NO₃⁻-N conversion with a superior ammonia selectivity of 99.60% at −1.2 V. Simultaneously, with the assistance of electrochlorination, the electrocatalytic system efficiently converts NH₄⁺-N to N₂, achieving a final selectivity of 99.41%. The on-line DEMS was carried out to capture the intermediates for inferring the reaction pathway of NO₃⁻RR, and DFT calculations were performed based on the pathway. It is found that re-Cu NRs with amorphous boundaries modulate the adsorption energy of reactants/intermediates and effectively inhibit the occurrence of HER, leading to excellent electrocatalytic activity and selectivity. This work proves that the amorphous Cu-based catalysts prepared with electrochemical in-situ reconfiguration exhibit high efficiency and stability in electrocatalytic NO₃⁻RR performance.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Yunqing Zhu: Conceptualization, Methodology, Writing – review & editing, Supervision, Project administration, Funding acquisition, Resources, Validation. Kaiyue Wen: Data curation, Conceptualization, Methodology, Validation, Investigation, Writing – original draft. Xuequan Wan: Methodology, Validation, Investigation. Gaigai Dong: Supervision, Validation, Visualization. Junfeng Niu: Supervision, Formal analysis, Methodology, Supervision, Resources.

This work was financially supported by the National Natural Science Foundation of China (No. 21876105), Shaanxi "Scientist & Engineer" Team (No. 2023KXJ-131) and Xianyang Key S & T Special Projects (No. L2023-ZDKJ-QCY-SXGG-GY-007). The authors would like to thank Shiyanjia Lab (www.shiyanjia.com) for supporting SEM and XPS tests.

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110399.