The copper source, copper(II) nitrate trihydrate (Cu(NO₃)₂·3H₂O, ≥ 99% purity), was purchased from Aladdin (China). The nitrogen/carbon sources, 2-methylimidazole and zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, both ≥ 99% purity), were also purchased from Aladdin (China). Sodium chloride (NaCl, analytical grade) was used as the chlorine source and purchased from Sinopharm Chemical Reagent (China). Methanol (CH₃OH, ≥ 99.5%) and ethanol (C₂H₅OH, ≥ 99.7%), both of analytical grade, were purchased from Macklin (China), while deionized water was prepared in the laboratory. Nafion solution (5 wt%) was purchased from Sigma-Aldrich (USA), and commercial Pt/C (20 wt%) was purchased from Macklin (China). Potassium hydroxide (KOH, ≥ 99.99% purity) used as the electrolyte was purchased from Aladdin (China). All aqueous solutions were prepared using 18.2 MΩ·cm Millipore water purified in a Millipore system. All chemicals were of analytical grade and used without further purification.

9.5 g of Zn(NO₃)₂·6H₂O and 20 g of 2-methylimidazole were dissolved in 200 mL of methanol, respectively. The reaction mixture was then stirred at room temperature for 24 h. The precipitate was collected by centrifugation at 6,000 rpm for 6 min, and the resulting solid was washed three times with methanol, and dried in a vacuum oven at 70 °C for about 12 h.

The ZIF-8 powder prepared above was ground and then high temperature pyrolysis. Under the protection of nitrogen (N₂) atmosphere, the temperature of the furnace was increased to 1,000 °C at a rate of 5 °C /min and maintained at a constant temperature for 2 h.

0.8 g of ZIF-8 was homogeneously dispersed in 100 mL of methanol, followed by the slow dropwise addition of a methanolic solution of 1 mmol of Cu(NO₃)₂·3H₂O to the above system, and magnetic stirring (500 rpm) for 6 h at room temperature (25 ± 1 °C). After centrifugation (8,000 rpm, 2 min), vacuum dried at 70 °C for 12 h. The resulting precursor was pyrolyzed in a nitrogen atmosphere at 5 °C/min up to 1,000 °C for 2 h.

The Ax-Cl-Cu/NC catalyst was prepared by mechanical milling combined with high-temperature pyrolysis: Cu/NC was fully milled and mixed with NaCl at a mass ratio of 1:1, heated to 1,000 °C at an elevated rate of 5 °C /min under nitrogen atmosphere and held for 2 h. In addition, we prepared samples with different NaCl to Cu ratios (0.75:1 and 1.25:1) by grinding for comparison. The product was washed with distilled water three times (30 mL each time) to completely remove the residual NaCl, followed by vacuum drying at 70 °C for 12 h.

The additional detailed experimental data, such as reagents, characterizations, and electrochemical measurements, are provided in the Supplementary Materials.

Ax-Cl-Cu/NC catalyst in which coordination-regulated Cu sites were confined in N-doped carbon carrier was prepared via a facile pyrolysis strategy. As seen in Figure 1A, ZIF-8 was prepared using a common method, and then the copper source (Cu precursor) and chlorine source (Cl precursor) were anchored to the ZIF-8 substrate surface using injection and grinding methods^[7,32]. Use a microinjection to drop metallic copper into the ZIF-8 solution, allowing Cu to be uniformly and stably anchored on the substrate. Grind NaCl and Cu-ZIF in a 1:1 ratio so that the Cl source can coordinate with Cu along the axial direction. The final sample is named Ax-Cl-Cu/NC catalyst and Cu/NC samples without sodium chloride were used as controls. Analyze the morphology of the sample using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Supplementary Figure 1A and B, both Ax-Cl-Cu/NC and Cu/NC exhibit a rhombic dodecahedral morphology similar to that of ZIF-8. High-resolution TEM (HRTEM) analysis showed [Supplementary Figure 1C and D] that no discernible lattice fringes were observed in either the Ax-Cl-Cu/NC or Cu/NC samples. X-ray diffraction (XRD) patterns further corroborate this conclusion. Only two distinct carbon diffraction peaks appear in the diffraction patterns of all samples [Supplementary Figure 2A] and XRD test results of samples at different temperatures [Supplementary Figure 2B] indicate that the form of Cu presence is not significantly related to temperature. Raman spectroscopy was further employed to examine the carbon structure. As displayed in Supplementary Figure 2C, both catalysts exhibit similar I_D/I_G ratios (≈1.001), indicating negligible change in graphitization degree upon axial Cl introduction. The specific morphology of Cu was further investigated using high-angle annular dark-field STEM (HAADF-STEM) [Figure 1B and C]. The results show that only isolated speckled bright spots (marked by red circles) appear in the Cu/NC and Ax-Cl-Cu/NC samples, indicating that metallic copper is atomically dispersed on the NC material.

The energy-dispersive X-ray spectroscopy (EDS) elemental distribution map of Ax-Cl-Cu/NC [Figure 1D] shows that carbon (C), nitrogen (N), copper (Cu), and chlorine (Cl) elements are uniformly distributed in the nitrogen carbon co-doped matrix, indicating the successful introduction of Cl. Further analysis of the coordination environment of Cu was performed using EXAFS spectroscopy [Figure 1E]. Compared with Cu foil and CuO, the prominent peak at around ~1.5 Å can be attributed to the Cu-N bond, while no Cu-Cu scattering peak appears in Ax-Cl-Cu/NC and Cu/NC. Interestingly, the Cu-N peak in Ax-Cl-Cu/NC shifted positively by 0.06 Å compared to the Cu/NC sample, indicating the formation of a Cu-Cl coordination structure. In addition, the fitted curves of the k³-weighted Cu K-edge EXAFS spectra in Figure 1E, as well as the corresponding data listed in Supplementary Table 1 for Ax-Cl-Cu/NC and Cu/NC, indicate that the coordination number (CN) of the Cu-N bond in Cu/NC is 3.9. For Ax-Cl-Cu/NC, the additional Cu-Cl pathway was taken into account during the fitting process, containing 3.9 Cu-N bonds and 1.0 Cu-Cl bonds. To highlight the local structure around the Cu site and the axial Cl coordination, a local structure model is shown in Figure 1E. These results demonstrate that the Cl element is indeed involved in the coordination of Cu-N₄ in an axial fashion. Quantitative analysis by inductively coupled plasma optical emission spectroscopy (ICP-OES) showed that the copper loading of Ax-Cl-Cu/NC and Cu/NC was 1.1 and 1.4 wt%, respectively.

The chemical composition and structure of each element in the catalyst were examined systematically. Firstly, the chemical composition of the catalyst was characterized using XPS. The full spectrum analysis in Supplementary Figure 3A shows the Cl characteristic peak, indicating the successful doping of Cl element in the Ax-Cl-Cu/NC catalyst. The C 1s XPS spectra of different samples are presented in Supplementary Figures 3B-D. All samples show a deconvoluted peak at 284.8 eV corresponding to C-N bonds, indicating that carbon and nitrogen are indispensable components of the catalysts. The local electronic states of each atom in the Ax-Cl-Cu/NC catalyst were further investigated by soft X-ray absorption near-edge structure spectroscopy (XANES) [Figure 2A], we can see that Cu/NC and Ax-Cl-Cu/NC have peaks at ~401 eV, which are attributed to Cu-N coordination. The regulatory effect of Cl atoms on Cu-N bonds was further revealed by XPS of N 1s, as shown in Figure 2B. It is worth noting that the binding energy of Cu-N bonds in Ax-Cl-Cu/NC shifted approximately 0.2 eV to the high energy compared to Cu/NC. Regarding the L-edge of Cl in the Ax-Cl-Cu/NC catalyst [Figure 2C], we can see two large peaks, G and H, which correspond to Cl-metal peaks, proving that Cl interacts with metal Cu^[33,34]. In the high-resolution Cl 2p spectrum of Ax-Cl-Cu/NC [Supplementary Figure 4], the Cu-Cl characteristic peak at 197.88 eV, directly confirming the presence of Cu-Cl bonds in the catalyst. The Cu 2p fine spectrum [Figure 2D] clearly shows that the peak of Cu²⁺ in Ax-Cl-Cu/NC shifts to a higher binding energy by 0.23 eV. This is because Cl has extremely high electronegativity and can attract electrons from around copper, leading to an increase in the valence state of Cu. Collectively, these findings suggest that the introduction of axial Cl induces a redistribution of d-band electrons around Cu, shifting electron density overall toward Cl, breaking the local electron symmetry around Cu, and thus potentially enhancing its ORR performance. The Cu K-edge XANES spectrum [Figure 2E] shows that the absorption energy of Ax-Cl-Cu/NC lies between that of Cu foil and CuO, close to that of +2 valent copper, and its absorption edge is slightly shifted to the high energy compared to Cu/NC, indicating that Cl coordination causes electron transfer from Cu to Cl, thereby increasing the apparent valence state of Cu. In the inset, the pre-edge peak Ax-Cl-Cu/NC shows a distinct small peak compared to Cu/NC, suggesting that the presence of Cl disrupts the symmetric configuration of CuN₄. Wavelet transform (WT)-EXAFS analysis further supports the above conclusion [Figure 2F]. The WT-EXAFS spectrum of Ax-Cl-Cu/NC shows slight differences compared to Cu/NC, indicating specific interactions between Cu and Cl atoms. Some fitting plots related to XAFS are provided in the supplementary information [Supplementary Figures 5 and 6]. The above X-ray absorption spectroscopy (XAS) results and XPS data form a complete chain of evidence, jointly confirming that Cl atoms the symmetry of electronic distribution through electron transfer and coordination effects.

Informed by the analysis of these results, the systematic evaluation of the ORR performance of the catalyst was carried out in an O₂-saturated 0.1 M KOH electrolyte using the RotatingRing-disk Electrode technique (DC DSR, PHYCHEMI)^[35]. As shown in Figure 3A, Ax-Cl-Cu/NC exhibited optimal ORR activity with a half-wave potential (E_1/2) of 0.89 V, which was significantly better than 20 wt% Pt/C (0.86 V), Cu/NC (0.85 V), and NC (0.63 V). Meanwhile, the ORR performance achieved by Ax-Cl-Cu/NC is superior to those of most recently reported SACs [Supplementary Table 2]. The experimental results of pyrolysis temperature optimization in Supplementary Figure 7 showed that the sample prepared by pyrolysis at 1,000 °C had the highest half-wave potential, which fully verified the key role of high-temperature treatment on the formation of active sites. We also measured the linear scan voltammogram curves of different NaCl-to-Cu/NC mass ratios, and the results confirmed that the 1:1 ratio gave the best performance [Supplementary Figure 8]. The onset potential of Ax-Cl-Cu/NC was 0.977 V (corresponding to 5% diffusion-limited current), which was slightly higher than that of Pt/C (0.973 V), and its limiting current density reached 6.12 mA cm^-2, which was significantly higher than that of Pt/C (5.02 mA cm^-2, Supplementary Figure 9), which indicates that the catalyst has better mass transfer efficiency for O₂. All catalysts were tested under the same conditions [Supplementary Figure 10]. As shown in Figure 3B, the kinetic current density (J_K) of Ax-Cl-Cu/NC at 0.85 V is as high as 17.399 mA cm^-2, which is 3.71 times higher than that of Pt/C (4.693 mA cm^-2), which directly confirms that the strengthening effect of Cl coordination on the CuN₄ active site. As can be seen from Figure 3C, the Tafel slope of Ax-Cl-Cu/NC was 109.92 mV dec^-1, which was significantly lower than that of Pt/C (131.2 mV dec^-1), Cu/NC (207.73 mV dec^-1), and NC (215.11 mV dec^-1), which fully proved its faster reaction kinetics. Ax-Cl-Cu/NC showed a turnover frequency (TOF) of 734.8 h^-1 at 0.85 V and a mass activity (MA) of 4,430.6 A g_metal^-1, which were 22.1 and 67.4 times higher than those of Pt/C (33.2 h^-1, 65.73 A g_metal^-1), respectively [Figure 3D]. This excellent performance is rooted in the induced effect of Cl atoms on Cu 3d energy-band electrons, which promotes the ORR kinetics. To understand the superior ORR activity of the Ax-Cl-Cu/NC catalyst, the electrochemical active surface areas (ECSA) of different catalysts were measured (ECSA = C_dl/C_s, where C_dl is the double-layer capacitance and Cs is the specific capacitance of graphitic carbon; here C_s was taken as 0.04 mF cm^-2)^[36]. Supplementary Figure 11A shows that among the catalysts prepared in this study, the Ax-Cl-Cu/NC catalyst exhibits the highest ECSA value (349 m² g^-1), outperforming Cu/NC (288 m² g^-1) and Pt/C (79 m² g^-1), confirming that this catalyst has the most active electron transfer centers during ORR. By normalizing the specific ORR activity with the ECSA values, the intrinsic activity of the Ax-Cl-Cu/NC catalyst was determined to be 0.0139 mA cm^-2, which is 3.18 times higher than that of the Cu/NC catalyst [Supplementary Figure 11B]. This indicates that the high intrinsic ORR activity of the Ax-Cl-Cu/NC catalyst originates from its unique Cl coordination environment. Electrochemical impedance spectroscopy (EIS) technique, and the results showed that Ax-Cl-Cu/NC has a lowest reaction resistance and thus exhibits a higher reactivity advantage [Supplementary Figure 12].

The electron transfer number (n) and H₂O₂ yield of the catalyst are key indicators of ORR selectivity^[37]. As measured by the rotating ring disk electrode (RRDE) technique [Figure 3E], the electron transfer number of Ax-Cl-Cu/NC is about 3.96, which is close to the ideal 4e^- transfer pathway, confirming its excellent ORR selectivity. The results of fitting the Koutechy-Levich (K-L) equation at different rotational speeds [Supplementary Figures 13 and 14] showed a high degree of consistency in the linear slopes, which further verified that the catalyst followed an efficient 4e^- reaction mechanism over a wide range of mass transfer. The H₂O₂ yield of Ax-Cl-Cu/NC was only 2.03 %, which was significantly lower than that of the other reference samples, suggesting that Cl doping inhibits the occurrence of side reactions. More importantly, this outstanding activity can remain stable even under rigorous testing conditions. Additionally, the stability and durability of the Ax-Cl-Cu/NC catalyst were measured via chronoamperometry (CA) and accelerated durability testing (ADT). ADT results showed [Figure 3F] that the half-wave potential (E_1/2) of Ax-Cl-Cu/NC only decays ~7 mV after long-term cycling, exhibiting a structural stability superior to that of most SACs. However, the E_1/2 of Pt/C and Cu/NC catalysts decreased by 18 and 11 mV, respectively [Supplementary Figure 15]. The Ax-Cl-Cu/NC exhibits reliable stability with better methanol resistance and lower current density attenuation (< 5%) after 40 h of CA testing at 0.7 V [Supplementary Figures 16 and 17]. Interestingly, the XPS of Cl in Ax-Cl-Cu/NC after the reaction shows no significant change compared to before the reaction, indicating its excellent stability [Supplementary Figure 4]. Post-durability XPS analysis shows no significant change in the Cl 2p peak position or intensity compared to the fresh sample, with the Cl atomic percentage remaining nearly unchanged (from 0.90% to 0.85%). This confirms that the axial Cl species do not undergo noticeable leaching or changes in chemical state under alkaline conditions. The experimental evidence indicates that the axial Cl in the Ax-Cl-Cu/NC catalyst exhibits good structural stability during alkaline ORR operation. By combining various characterizations and performance test results, the introduction of chlorine achieved a simultaneous enhancement of ORR selectivity and stability.

To elucidate the mechanism behind the outstanding ORR activity and stability of Ax-Cl-Cu/NC, in-situ EIS and in-situ synchronous radiation infrared spectroscopy (SRIR) technique studies were conducted. The distribution of relaxation times (DRT) of Cu/NC and Ax-Cl-Cu/NC at different voltages [Figure 4A and B] reveal two dominant characteristic peaks, corresponding to mass transport (10^-1~10⁰ Hz) and charge transport (~10¹ Hz) processes, respectively^[38-40]. Comparative analysis demonstrates that Ax-Cl-Cu/NC exhibits significantly reduced peak intensities with all peaks shifted toward higher frequencies compared to Cu/NC. Charge Transfer Resistance (R_ct) is related to the charge transfer at the electrode interface; the higher the R_ct, the weaker the charge transfer. Figure 4C shows that the reaction kinetics of Ax-Cl-Cu/NC are stronger than those of Cu/NC at any voltage. The EIS plots of the Ax-Cl-Cu/NC and Cu/NC catalysts at different voltages are shown in Supplementary Figure 18. These observations indicate that Ax-Cl-Cu/NC possesses lower charge transfer resistance, which are more favorable for facilitating the ORR process.

In-situ SRIR technique monitor the evolution of active species in the ORR process of Ax-Cl-Cu/NC and Cu/NC catalysts, and the regulatory mechanism of coordination Cl on the reaction path was revealed^[41]. As shown in Figure 4D and E, both catalysts were tested for potential from 1.0 to 0.4 V in O₂-saturated 0.1 M KOH^[42,43]. From the figure, we can see that the infrared spectrum of Cu/NC only shows *O (~900 cm^-1)^[44,45] and *OOH (~1,180 cm^-1)^[46,47] signals, whereas the infrared spectrum of Ax-Cl-Cu/NC also shows H-O-H (~1,630 cm^-1)^[48,49] signals, indicating that the latter more effectively promotes the cleavage of the O-O bond^[50-52]. Detailed peak assignments are provided in Supplementary Table 3. The resulting oxygen atoms can quickly combine with proton-electron pairs to form OH, which is further protonated to ultimately produce water (H-O-H). The cleavage of the O-O bond marks the shift of the reaction pathway from a 2e^- route to a more efficient 4e^- route. The potential-dependent behavior of oxygen intermediates was investigated by SRIR spectroscopy [Figure 4F]. Notably, Ax-Cl-Cu/NC showed much stronger *O signals than Cu/NC, attributed to its enhanced *OOH decomposition capability, facilitating rapid *O formation. In situ characterization results not only intuitively confirmed that this catalyst has excellent reaction kinetics. More importantly, by precisely detecting the key reaction intermediates (*O, *OOH) in the ORR, it directly revealed the intrinsic reasons for its high performance, providing experimental evidence for promoting the ORR process.

In order to verify the practical application value of Ax-Cl-Cu/NC catalyst, it was used as the cathode material to construct ZABs and compared with commercially available Pt/C catalysts^[53,54]. The ZAB was assembled using a polished zinc plate as the anode, Ax-Cl-Cu/NC loaded on carbon paper as the air cathode, and 6 M KOH as the electrolyte [Supplementary Figure 19]. All tests were conducted under high-purity O₂ (99.999%) atmosphere at a flow rate of 50 mL min^-1, eliminating CO₂ interference and minimizing carbonate formation in the electrolyte. As shown in Figure 5A, the Ax-Cl-Cu/NC equipped ZAB exhibited an OCV of 1.53 V, which was significantly higher than that of the Pt/C based battery of 1.44 V, indicating that the catalyst could drive higher theoretical output voltage of the battery. The discharge polarization curves and power density tests in Figure 5B showed that the Ax-Cl-Cu/NC based ZABs exhibited a maximum output power density of 156.6 mW cm^-2, higher than that of the Pt/C based cells at 127.3 mW cm^-2, and a superior discharge voltage plateau reflecting a more efficient oxygen reduction kinetics. The performance that surpasses most of the catalysts reported in the literature [Supplementary Table 4]. At a current density of 10 mA cm^-2, the specific capacity of the Ax-Cl-Cu/NC based ZAB was 751.1 mA h g_Zn^-1, which was significantly better than that of the Pt/C based cell (686.4 mA h g_Zn^-1) [Figure 5C], indicating a more efficient utilization of the zinc anode. The dynamic current density test [Figure 5D] showed that the Ax-Cl-Cu/NC based battery always maintained a higher discharge voltage than the Pt/C based battery when the current density was increased from 5 to 50 mA cm^-2. And the discharge potential quickly recovered to the initial level after the current density was reduced to 5 mA cm^-2, confirming its excellent multiplicative performance and reversibility. Cycling durability tests showed that the discharge voltage of Ax-Cl-Cu/NC based ZAB did not show any obvious decay after 100 h at a current density of 5 mA cm^-2, and the discharge-charge voltage difference remained stable during long-term charging and discharging [Figure 5E], which further verified the long-cycle durability of the catalyst under practical working conditions. However, the Pt/C-based ZAB showed significant performance degradation after 30 h [Supplementary Figure 20]. The above results fully demonstrate that the Ax-Cl-Cu/NC catalyst, with its excellent oxygen reduction performance and stable structural properties, exhibits significant potential for applications in energy storage fields such as zinc air batteries.