Thank you for visiting Nature.com. You are using a browser version with limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In addition, to ensure continued support, we show the site without styles and JavaScript.
        Sliders showing three articles per slide. Use the back and next buttons to move through the slides, or the slide controller buttons at the end to move through each slide.
        Efficient, inexpensive and durable oxygen reduction reaction (ORR) electrocatalysts are of great importance for secondary Zn-air batteries. The ORR activity of single and mixed metal oxides and carbon electrocatalysts was investigated using rotating disk electrode (RDE) measurements, Tafel slopes, and Kutetsky-Levich plots. It was found that the combination of MnOx and XC-72R exhibits high PBP activity and good stability, up to 100 mA cm–2. The performance of the selected ORR electrodes and the previously optimized oxygen evolution reaction (OER) electrode were then tested in a custom-built secondary zinc-air battery in a three-electrode configuration, and current density, electrolyte molarity, temperature, oxygen purity were also tested. Characteristics of ORR and OER electrodes. Finally, the durability of the secondary zinc-air system was evaluated, demonstrating an energy efficiency of 58–61% at 20 mA cm-2 in 4 M NaOH + 0.3 M ZnO at 333 K for 40 hours.
        Metal-air batteries with oxygen electrodes are considered extremely attractive systems because electroactive materials for oxygen electrodes can be easily obtained from the surrounding atmosphere and do not require storage1. This simplifies the system design by allowing the oxygen electrode to have unlimited capacity, thereby increasing the energy density of the system. Therefore, metal-air batteries using anode materials such as lithium, aluminum, iron, zinc, and magnesium have emerged due to their excellent specific capacity. Among them, zinc air batteries are quite capable of meeting the market demand for cost, safety, and environmental friendliness, since zinc has many desirable characteristics as an anode material, such as good stability in aqueous electrolytes, high energy density, and low equilibrium. potential., electrochemical reversibility, good electrical conductivity, abundance and ease of handling4,5. Currently, although primary zinc air batteries are used in commercial applications such as hearing aids, railway signals and navigation lights, secondary zinc air batteries have the potential for high energy density comparable to lithium-based batteries. This makes it worthwhile to continue research on zinc air batteries for applications in portable electronics, electric vehicles, grid-scale energy storage and to support renewable energy production6,7.
        One of the key objectives is to improve the efficiency of oxygen reactions at the air electrode, namely the oxygen reduction reaction (ORR) and the oxygen evolution reaction (OER), in order to promote the commercialization of secondary Zn-air batteries. To this end, efficient electrocatalysts can be used to increase the reaction rate and thus increase efficiency. At present, oxygen electrodes with bifunctional catalysts are well described in the literature8,9,10. Although bifunctional catalysts can simplify the structure of electrodes and reduce mass transfer losses, which can help reduce production costs, in practice, catalysts that are best suited for ORR are often not suitable for OER, and vice versa11. This difference in operating potential causes the catalyst to be exposed to a wider range of potentials, which can change its surface structure over time. In addition, the interdependence of intermediate binding energies means that active sites on the catalyst can be different for each reaction, which can complicate optimization.
        Another major problem for secondary Zn-air batteries is the design of the oxygen electrode, mainly because the monofunctional catalysts for ORR and OER operate in different reaction media. The ORR gas diffusion layer must be hydrophobic to allow oxygen gas to enter the catalytic sites, while for OER the electrode surface must be hydrophilic to facilitate the removal of oxygen bubbles. On fig. 1 shows three typical secondary oxygen electrode designs taken from a review by Jorissen12, namely (i) bifunctional monolayer catalysts, (ii) double or multilayer catalysts, and (iii) triple electrode configurations.
        For the first electrode design, which includes only a single layer bifunctional catalyst that simultaneously catalyzes ORR and OER, if a membrane is included in this design, then a membrane-electrode assembly (MEA) is formed as shown. The second type includes two (or more) catalyst beds with different porosity and hydrophobicity to account for differences in reaction zones13,14,15. In some cases, the two catalytic beds are separated, with the hydrophilic side of the OER facing the electrolyte and the semi-hydrophobic side of the ORR facing the open ends of the electrodes 16, 17, 18. a cell consisting of two reaction-specific oxygen electrodes and a zinc electrode19,20. Table S1 lists the advantages and disadvantages of each design.
        The implementation of an electrode design that separates the ORR and OER reactions has previously shown improved cycling stability19. This is especially true for the three electrode configuration, where degradation of unstable catalysts and co-additives is minimized and outgassing is more controllable over the entire potential range. For these reasons, we used a three-electrode Zn-air configuration in this work.
        In this article, we first select high performance ORR catalysts by comparing various transition metal oxides, carbonaceous materials, and reference catalysts with rotating disk electrode (RDE) experiments. Transition metal oxides tend to be good electrocatalysts due to their varying oxidation states; reactions are more easily catalyzed in the presence of these compounds21. For example, manganese oxides, cobalt oxides, and cobalt-based mixed oxides (such as NiCo2O4 and MnCo2O4)22,23,24 show good ORR in alkaline conditions due to their half-filled d-orbitals, electron energy levels that allow for electron work and improved cutting comfort. In addition, they are more abundant in the environment and have acceptable electrical conductivity, high reactivity and good stability. Similarly, carbonaceous materials are widely used, having the advantages of high electrical conductivity and large surface area. In some cases, heteroatoms such as nitrogen, boron, phosphorus, and sulfur have been introduced into carbon to modify its structure, further improving the ORR characteristics of these materials.
        Based on the experimental results, we included the selected OVR catalysts in gas diffusion electrodes (GDE) and tested them at various current densities. The most efficient ORR GDE catalyst was then assembled into our custom three-electrode secondary Zn-air battery along with reaction-specific OER electrodes optimized in our previous work26,27. The potentials of individual oxygen electrodes were monitored during continuous discharge and charge cycling experiments to study the effect of operating conditions such as current density, electrolyte molarity, cell operating temperature, and oxygen purity. Finally, the stability of Zn-air secondary batteries was evaluated under continuous cycling under optimum operating conditions.
        MnOx28 was prepared by the chemical redox method: 50 ml of 0.04 M KMnO4 solution (Fisher Scientific, 99%) was added to 100 ml of 0.03 M Mn(CH3COO)2 (Fisher Scientific, 98%) to form a brown precipitate. The mixture is adjusted to pH 12 with dilute sodium hydroxide, then centrifuged 3-5 times at 2500 rpm to collect the precipitate. The precipitate was then washed with deionized water until the purple color of the permanganate ion disappeared. Finally, the deposits were air-dried at 333 K overnight and then pulverized.
        Spinel oxides Co3O4, NiCo2O4, and MnCo2O4 were synthesized by thermal decomposition. NiCo2O4 and MnCo2O4 were prepared by adding 0.5 M (14.5 g) nickel(II) nitrate hexahydrate, Ni(NO3)2⋅6H2O (Fisher Scientific, 99.9%) or 0.5 M (12.6 g) tetrahydrate manganese(II) nitrate Mn(NO3). )2 4H2O (Sigma Aldrich, ≥ 97%) and 1 M (29.1 g) cobalt(II) nitrate hexahydrate, Co(NO3)2 6H2O (Fisher Scientific, 98+%, ACS reagents) in methanol (Fisher Scientific , 99.9% ) in 100 ml dilution vials. Methanol is added in small portions to the transition metal nitrate with continuous stirring until a homogeneous solution is obtained. The solution was then transferred to a crucible and heated on a hot plate, leaving a dark red solid. The solid was calcined at 648 K for 20 h in air. The resulting solid was then ground to a fine powder. No Ni(NO3)2 6H2O or Mn(NO3)2 4H2O was added during the synthesis of Co3O4.
       Graphene nanosheets with a surface area of ​​300 m2/g (Sigma Aldrich), graphene doped with nitrogen (Sigma Aldrich), carbon black powder (Vulcan XC-72R, Cabot Corp., 100%), MnO2 (Sigma Aldrich) and 5 wt.% Pt/C (Acros Organics) was used as is.
        RDE (Pine Research Instrumentation) measurements were used to evaluate the activity of various ORR catalysts in 1 M NaOH. A catalytic ink consisting of 1 mg catalyst + 1 ml deionized (DI) H2O + 0.5 ml isopropanol (IPA) + 5 µl 5 wt% Nafion 117 (Sigma-Aldrich) was used as is. When Vulcan XC-72R was added, the catalytic paint consisted of 0.5 mg catalyst + 0.5 mg Vulcan XC-72R + 1 ml DI HO + 0.5 ml IPA + 5 µl 5 wt% Nafion 117 to ensure consistent material loading . The mixture was sonicated for 20 minutes and homogenized using a Cole-Parmer LabGen 7 Series homogenizer at 28,000 rpm for 4 minutes. The ink was then applied in three aliquots of 8 μl to the surface of a glassy carbon electrode (Pine Instrument Company) with a diameter of 4 mm (working area ≈ 0.126 cm2) and dried between layers to provide a load of ≈120 μg cm-2. Between applications, the glassy carbon electrode surface was successively wet polished with MicroCloth (Buehler) and 1.0 mm and 0.5 mm alumina powder (MicroPolish, Buehler) followed by sonication in deionized H2O.
        ORR gas diffusion electrode samples were prepared according to our previously described protocol28. First, the catalyst powder and Vulcan XC-72R were mixed in a 1:1 weight ratio. Then a mixture of a solution of polytetrafluoroethylene (PTFE) (60 wt.% in H2O) and a solvent with a ratio of IPA/H2O of 1:1 was added to the dry powder mixture. Sonicate the catalytic paint for about 20 minutes and homogenize for about 4 minutes at 28,000 rpm. The ink was then thinly applied with a spatula onto pre-cut carbon paper 13 mm in diameter (AvCarb GDS 1120) and dried until a catalyst content of 2 mg cm2 was reached.
        OER electrodes were fabricated by cathodic electrodeposition of Ni—Fe hydroxide catalysts onto a 15 mm x 15 mm stainless steel mesh (DeXmet Corp, 4SS 5-050) as reported26,27. Electrodeposition was carried out in a standard three-electrode half-cell (a polymer-coated glass cell of approximately 20 cm3) with a Pt grid as a counter electrode and Hg/HgO in 1 M NaOH as a reference electrode. Allow the catalyst coated stainless steel mesh to air dry before cutting out an area of ​​approximately 0.8 cm2 with a 10 mm thick carbon steel punch.
        For comparison, commercial ORR and OER electrodes were used as received and tested under the same conditions. The commercial ORR electrode (QSI Nano Gas Diffusion Electrode, Quantum Sphere, 0.35 mm thick) consists of manganese and carbon oxide coated with a nickel mesh current collector, while the commercial OER electrode (type 1.7, special Magneto anode, BV) has thickness 1.3 mm. up to 1.6 mm expanded titanium mesh coated with Ru-Ir mixed metal oxide.
        The surface morphology and composition of the catalysts were characterized using an FEI Quanta 650 FEG scanning electron microscope (SEM) operating under high vacuum and an accelerating voltage of 5 kV. Powder X-ray diffraction (XRD) data were collected on a Bruker D8 Advance X-ray diffractometer with a copper tube source (λ = 1.5418 Å) and analyzed using Bruker Diffraction Suite EVA software.
        All electrochemical measurements were performed using a Biologic SP-150 potentiostat and EC-lab software. Samples of RDE and GDE were tested on a standard three-electrode setup consisting of a 200 cm3 jacketed glass cell and a Laggin capillary as a reference electrode. Pt mesh and Hg/HgO in 1 M NaOH were used as counter and reference electrodes, respectively.
        For RDE measurements in each experiment, fresh 1 M NaOH electrolyte was used, the temperature of which was kept constant at 298 K using a circulating water bath (TC120, Grant). Gaseous oxygen (BOC) was bubbling into the electrolyte through a glass frit with a porosity of 25–50 µm for at least 30 min before each experiment. To obtain ORR polarization curves, the potential was scanned from 0.1 to -0.5 V (relative to Hg/HgO) at a scan rate of 5 mV s -1 at 400 rpm. Cyclic voltammograms were obtained by sweeping the potential between 0 and -1.0 V and Hg/HgO at a rate of 50 mV s-1.
        For HDE measurements, the 1 M NaOH electrolyte was maintained at 333 K with a circulating water bath. An active area of ​​0.8 cm2 was exposed to the electrolyte with a continuous supply of oxygen to the rear side of the electrode at a rate of 200 cm3/min. The fixed distance between the working electrode and the reference electrode was 10 mm, and the distance between the working electrode and the counter electrode was 13-15 mm. Nickel wire and mesh provide electrical contact on the gas side. Chronopotentiometric measurements were taken at 10, 20, 50 and 100 mA cm-2 to evaluate the stability and efficiency of the electrode.
        The characteristics of the ORR and OER electrodes were evaluated in a 200 cm3 jacketed glass cell with a PTFE29 insert. A schematic diagram of the system is shown in Figure S1. The electrodes in the battery are connected in a three-electrode system. The working electrode consisted of separate reaction-specific ORR and OER electrodes connected to a relay module (Songle, SRD-05VDC-SL-C) and a microcontroller (Raspberry Pi 2014© model B+V1.2) with a zinc anode. as a pair The electrodes and the reference electrode Hg/HgO in 4 M NaOH were at a distance of 3 mm from the zinc anode. A Python script has been written to operate and control the Raspberry Pi and Relay Module.
        The cell was modified to accommodate a zinc foil anode (Goodfellow, 1 mm thick, 99.95%) and a polymer cover allowed the electrodes to be placed at a fixed distance of approximately 10 m. 4 mm apart. Nitrile rubber plugs fixed the electrodes in the lid, and nickel wires (Alfa Aesar, 0.5 mm diameter, annealed, 99.5% Ni) were used for the electrical contacts of the electrodes. The zinc foil anode was first cleaned with isopropanol and then with deionized water, and the surface of the foil was covered with polypropylene tape (Avon, AVN9811060K, 25 µm thick) to expose an active area of ​​approximately 0.8 cm2.
        All cycling experiments were performed in 4 M NaOH + 0.3 M ZnO electrolyte at 333 K unless otherwise noted. In the figure, Ewe with respect to Hg/HgO refers to the potential of the oxygen electrode (ORR and OER), Ece with respect to Hg/HgO represents the potential of the zinc electrode, Ecell with respect to Hg/HgO represents the full cell potential or potential difference. between two battery potentials. Oxygen or compressed air was supplied to the rear side of the OPP electrode at a constant flow rate of 200 cm3/min. The cycling stability and performance of the electrodes was studied at a current density of 20 mA cm-2, a cycle time of 30 min, and an OCV rest time of 1 min between each half cycle. A minimum of 10 cycles were performed for each test, and data was extracted from cycles 1, 5, and 10 to determine the condition of the electrodes over time.
        The morphology of the ORR catalyst was characterized by SEM (Fig. 2), and powder X-ray diffraction measurements confirmed the crystal structure of the samples (Fig. 3). The structural parameters of the catalyst samples are given in Table 1. 1. When comparing manganese oxides, commercial MnO2 in fig. 2a consists of large particles, and the diffraction pattern in Fig. 3a corresponds to JCPDS 24-0735 for tetragonal β-MnO2. On the contrary, on the MnOx surface in Fig. 2b shows finer and finer particles, which corresponds to the diffraction pattern in Fig. 66° correspond to the peaks (110), (220), (310), (211), and (541) of the tetrahedrally centered α-MnO2 hydrate, JCPDS 44-014028.
       (a) MnO2, (b) MnOx, (c) Co3O4, (d) NiCo2O4, (e) MnCo2O4, (f) Vulcan XC-72R, (g) graphene, (h) nitrogen doped graphene, (and ) 5 wt.% Pt/C.
       X-ray patterns of (a) MnO2, (b) MnOx, (c) Co3O4, (d) NiCo2O4, (e) MnCo2O4, (f) Vulcan XC-72R, nitrogen-doped graphene and graphene, and (g) 5% platinum /carbon.
        On fig. 2c–e, the surface morphology of oxides based on cobalt Co3O4, NiCo2O4, and MnCo2O4 consists of clusters of irregularly sized particles. On fig. 3c–e show that all of these transition metal oxides have a spinel structure and a similar cubic crystal system (JCPDS 01-1152, JCPDS 20-0781, and JCPDS 23-1237, respectively). This indicates that the thermal decomposition method is capable of producing highly crystalline metal oxides, as evidenced by the strong well-defined peaks in the diffraction pattern.
        SEM images of carbon materials show large changes. On fig. 2f Vulcan XC-72R carbon black consists of densely packed nanoparticles. On the contrary, the appearance of graphene in Fig. 2g are highly disordered plates with some agglomerations. However, N-doped graphene (Fig. 2h) appears to consist of thin layers. The corresponding X-ray diffraction patterns of Vulcan XC-72R, commercial graphene nanosheets, and N-doped graphene in Figs. 3f show small changes in the 2θ values ​​of the (002) and (100) carbon peaks. Vulcan XC-72R is identified as a hexagonal graphite in JCPDS 41-1487 with peaks (002) and (100) appearing at 24.5° and 43.2° respectively. Similarly, the (002) and (100) peaks of N-doped graphene appear at 26.7° and 43.3°, respectively. The background intensity observed in the X-ray diffraction patterns of Vulcan XC-72R and nitrogen-doped graphene is due to the highly disordered nature of these materials in their surface morphology. In contrast, the diffraction pattern of graphene nanosheets shows a sharp, intense peak (002) at 26.5° and a small broad peak (100) at 44°, indicating a more crystalline nature of this sample.
        Finally, in fig. 2i SEM image of 5 wt.% Pt/C shows rod-shaped carbon fragments with round voids. Cubic Pt is determined from most of the peaks in the 5 wt% Pt/C diffraction pattern in Fig. 3g, and the peak at 23° corresponds to the (002) peak of the carbon present.
        A linear sweep ORR catalyst voltammogram was recorded at a sweep rate of 5 mV s-1. Due to mass transfer limitations, the collected maps (Fig. 4a) usually have an S-shape extending to a plateau with more negative potential. The limiting current density, jL, potential E1/2 (where j/jL = ½) and onset potential at -0.1 mA cm-2 were extracted from these plots and listed in Table 2. It is worth noting that in fig. 4a, catalysts can be classified according to their E1/2 potentials into: (I) metal oxides, (II) carbonaceous materials, and (III) noble metals.
       Linear sweep voltammograms of (a) catalyst and (b) a thin film of catalyst and XC-72R, measured on an RDE glassy carbon probe at 400 rpm with a scan rate of 5 mV s-1 in O2 saturation at 298 K in 1 M NaOH cf.
        Individual metal oxides of Mn and Co in group I show initial potentials of -0.17 V and -0.19 V respectively, and E1/2 values ​​are between -0.24 and -0.26 V. The reduction reactions of these metal oxides are presented in equation. (1) and (2), which appear next to the onset potential in Figs. 4a match the standard potential of the first step 2e of the ORR indirect path in the equation. (3).
       The mixed metal oxides MnCo2O4 and NiCo2O4 in the same group show slightly corrected initial potentials at -0.10 and -0.12 V respectively, but retain E1/2 values ​​of about 10.−0.23 volts.
        Group II carbon materials show more positive E1/2 values ​​than group I metal oxides. Graphene material has an initial potential of -0.07 V and an E1/2 value of -0.11 V, while an initial potential and E1/2 of 72R Vulcan XC- are -0.12V and -0.17V respectively. In group III, 5 wt% Pt/C showed the most positive initial potential at 0.02 V, an E1/2 of -0.055 V, and a maximum limit at -0.4 V, since oxygen reduction occurred via the current density of the 4e path . It also has the lowest E1/2 due to the high conductivity of Pt/C and the reversible kinetics of the ORR reaction.
        Figure S2a presents the Tafel slope analysis for various catalysts. The kinetically controlled region of 5 wt.% Pt/C starts at 0.02 V with respect to Hg/HgO, while the region of metal oxides and carbon materials is in the range of negative potentials from -0.03 to -0.1 V. The slope value for Tafel Pt/C is –63.5 mV s s–1, which is typical for Pt at low current densities dE/d log i = –2.3 RT/F31.32 in which the rate-determining step involves the transition of oxygen from physisorption to chemisorption33,34. The Tafel slope values ​​for carbon materials are in the same region as Pt/C (-60 to -70 mV div-1), suggesting that these materials have similar ORR paths. Individual metal oxides of Co and Mn report Tafel slopes ranging from -110 to -120 mV dec-1, which is dE/d log i = -2.3 2RT/F, where the rate-determining step is the first electron. transfer step 35, 36. Slightly higher slope values ​​recorded for mixed metal oxides NiCo2O4 and MnCo2O4, about -170 mV dec-1, indicate the presence of OH- and H2O ions on the surface of the oxide, which prevent oxygen adsorption and electron transfer, thereby affecting oxygen. reduction path 35.
        The Kutetsky-Levich (KL) equation was used to determine the kinetic reaction parameters for various catalyst samples without mass transfer. in the equation. (4) the total measured current density j is the sum of the current densities of electron transfer and mass transfer.
        from the equation. (5) The limiting current density jL is proportional to the square root of the rotation speed. Therefore, the KL equation. (6) describes a line graph of j−1 versus ω−1//2, where the intersection point is jk and the slope of the graph is K.
       where ν is the kinematic viscosity of the electrolyte 1 M NaOH (1.1 × 10–2 cm2 s–1)37, D is the diffusion coefficient of O2 in 1 M NaOH (1.89 × 10–5 cm2 s–1)38, ω is rpm is the rotation speed, C is the oxygen concentration in the bulk solution (8.4 × 10–7 mol cm–3)38.
        Collect linearly swept voltammograms using RDE at 100, 400, 900, 1600, and 2500 rpm. Values ​​were taken from -0.4 V in the limited mass transfer region to plot the KL diagram, i.e. -j-1 versus ω-1//2 for the catalyst (Fig. S3a). Use equations. In equations (6) and (7), the performance indicators of the catalyst, such as the kinetic current density without taking into account the effects of mass transfer jk, are determined by the point of intersection with the y axis, and the number of electron transfers is determined by the gradient K of the curve. They are listed in table 2.
        5 wt% Pt/C and XC-72R have the lowest absolute jk values, indicating faster kinetics for these materials. However, the slope of the XC-72R curve is almost twice that for 5 wt% Pt/C, which is expected since K is an indication of the number of electrons transferred during the oxygen reduction reaction. Theoretically, the KL plot for 5 wt% Pt/C should pass through the 39 origin under limited mass transfer conditions, however this is not observed in Figure S3a, suggesting kinetic or diffusional limitations affecting the results. This may be because Garsani et al. 40 have shown that small inconsistencies in the topology and morphology of Pt/C catalytic films can affect the accuracy of ORR activity values. However, since all catalyst films were prepared in the same way, any effect on the results should be the same for all samples. The graphene KL cross point of ≈ -0.13 mA-1 cm2 is comparable to that of the XC-72R, but the -0.20 mA-1 cm2 cross point for the N-doped graphene KL graph indicates that the current density is greater depends on the voltage on the catalytic converter. This may be due to the fact that nitrogen doping of graphene reduces the overall electrical conductivity, resulting in slower electron transfer kinetics. In contrast, the absolute K value of nitrogen-doped graphene is smaller than that of graphene because the presence of nitrogen helps create more active sites for ORR41,42.
        For oxides based on manganese, the intersection point of the largest absolute value is observed – 0.57 mA-1 cm2. Nevertheless, the absolute K value of MnOx is much lower than that of MnO2 and is close to 5 wt %. %Pt/C. The electron transfer numbers were determined to be approx. MnOx is 4 and MnO2 is close to 2. This is consistent with results published in the literature, which report that the number of electron transfers in the α-MnO2 ORR path is 4, while β-MnO243 is typically less than 4. Thus Thus, the ORR pathways differ for different polymorphic forms of catalysts based on manganese oxide, although the rates of chemical steps remain approximately the same. In particular, the MnOx and MnCo2O4 catalysts have electron transfer numbers slightly higher than 4 because the reduction of manganese oxides present in these catalysts occurs simultaneously with the reduction of oxygen. In a previous work, we found that the electrochemical reduction of manganese oxide occurs in the same potential range as the reduction of oxygen in a solution saturated with nitrogen28. The contribution of side reactions leads to a calculated number of electrons slightly more than 4.
       The intersection of Co3O4 is ≈ −0.48 mA-1 cm2, which is less negative than the two forms of manganese oxide, and the apparent electron transfer number is determined by the value of K equal to 2. Replacing Ni in NiCo2O4 and Mn in MnCo2O4 by Co leads to a decrease in the absolute values K, which indicates an improvement in the electron transfer kinetics in mixed metal oxides.
        Carbon substrates are added to the ORR catalyst ink to increase electrical conductivity and facilitate proper three-phase boundary formation in gas diffusion electrodes. Vulcan-XC-72R was selected due to its low price, large surface area of ​​250 m2·g-1, and low resistivity of 0.08 to 1 Ω·cm44.45. An LSV plot of a catalyst sample mixed with Vulcan XC-72R at 400 rpm is shown in Figure 1. 4b. The most obvious effect of adding the Vulcan XC-72R is to increase the ultimate current density. Note that this is more noticeable for metal oxides, with an additional 0.60 mA cm-2 for single metal oxides, 0.40 mA cm-2 for mixed metal oxides, and 0.28 mA cm-2 for graphene and doped graphene. N. Add 0.05 mA cm-2. −2. The addition of Vulcan XC-72R to the catalyst ink also resulted in a positive shift in the onset potential and the E1/2 half-wave potential for all catalysts except graphene. These changes may be a possible result of increased electrochemical surface area utilization46 and improved contact47 between catalyst particles on supported Vulcan XC-72R catalyst.
        The corresponding Tafel plots and kinetic parameters for these catalyst mixtures are shown in Figure S2b and Table 3, respectively. The Tafel slope values ​​were the same for the MnOx and graphene materials with and without XC-72R, indicating that their ORR pathways were not affected. However, the cobalt-based oxides Co3O4, NiCo2O4 and MnCo2O4 gave smaller negative Tafel slope values ​​between -68 and -80 mV dec-1 in combination with XC-72R indicating a shift in the ORR pathway. Figure S3b shows a KL plot for a catalyst sample combined with a Vulcan XC-72R. In general, a decrease in the absolute values ​​of jk was observed for all catalysts mixed with XC-72R. MnOx showed the largest decrease in the absolute value of jk by 55 mA-1 cm2, while NiCo2O4 recorded a decrease by 32 mA-1 cm-2, and graphene showed the smallest decrease by 5 mA-1 cm2. It can be concluded that the effect of Vulcan XC-72R on the performance of the catalyst is limited by the initial activity of the catalyst in terms of OVR.
        Vulcan XC-72R does not affect the K values ​​of NiCo2O4, MnCo2O4, graphene, and nitrogen-doped graphene. However, the K value of Co3O4 decreased significantly with the addition of Vulcan XC-72R, indicating an increase in the number of electrons transferred by the ORR. Such co-association of Co3O4 with carbon components has been reported in refs. 48, 49. In the absence of a carbon support, Co3O4 is thought to promote disproportionation of HO2- to O2 and OH-50.51, which is in good agreement with Co3O4′s electron transfer number of about 2 in Table 2. Thus, the physical adsorption of Co3O4 on carbon substrates is expected to generate a 2 + 2 four-electron ORR pathway52 that first electroreduces O2 to HO2- at the interface of the Co3O4 catalyst and Vulcan XC-72R (equation 1) and then HO2 – The rapidly disproportionated metal oxide surface is converted to O2 followed by electroreduction.
        In contrast, the absolute value of K MnOx increased with the addition of Vulcan XC-72R, which represents a decrease in the electron transfer number from 4.6 to 3.3 (Table 3). This is due to the presence of two sites on the carbon catalyst composite for the two-stage electron path. The initial reduction of O2 to HO2- occurs more easily on carbon supports, resulting in a slightly increased preference for the two-electron pathway of ORR53.
        The stability of the catalyst was evaluated in the GDE half-cell in the range of current densities. On fig. 5 shows plots of potential versus time for GDE MnOx, MnCo2O4, NiCo2O4, graphene, and nitrogen-doped graphene. MnOx shows good overall stability and ORR performance at low and high current densities, suggesting that it is suitable for further optimization.
       Chronopotentiometry of HDE samples at current from 10 to 100 mA/cm2 in 1 M NaOH, 333 K, O2 flow rate 200 cm3/min.
        MnCo2O4 also appears to retain good ORR stability across the current density range, but at higher current densities of 50 and 100 mA cm-2 large overvoltages are observed indicating that MnCo2O4 does not perform as well as MnOx. Graphene GDE exhibits the lowest ORR performance over the current density range tested, demonstrating a rapid drop in performance at 100 mA cm-2. Therefore, under the chosen experimental conditions, MnOx GDE was chosen for further tests in the Zn-air secondary system.