The Role of the A-Site Cation on the Bifunctional Electrocatalytic Activities of Ln_0.5Sr_0.5CoO_3-δ (Ln = La, Pr and Sm) for Rechargeable Zinc–Air Batteries

Abstract

Developing efficient bifunctional electrocatalysts for both the oxygen reduction reaction (ORR) and the oxygen evolution reaction (OER) is crucial for the large-scale application of rechargeable zinc–air batteries. Perovskite oxides are attractive candidates as bifunctional electrocatalysts. Herein, Ln_0.5Sr_0.5CoO_3-δ (Ln = La, Pr and Sm) was prepared by the sol-gel method and evaluated as bifunctional electrocatalysts for both ORR and OER. Compare with La_0.5Sr_0.5CoO_3-δ (LSC) and Sm_0.5Sr_0.5CoO_3-δ (SSC), Pr_0.5Sr_0.5CoO_3-δ (PSC) demonstrated the enhanced catalytic activity towards ORR with the limiting current density of 4.19 mA cm⁻² at 0.40 V vs. RHE and OER with the potential of 1.75 V vs. RHE at 10 mA cm⁻² as well as the Tafel slope of 105 mV dec⁻¹. The higher Co³⁺/Co²⁺ ratio in PSC played an essential part in boosting the bifunctional electrocatalytic activities for ORR and OER. For demonstration, rechargeable zinc–air batteries with PSC as the air electrode displayed the maximum power density of 72 mW cm⁻², the low charge-discharge voltage gap (1.01 V) at 10 mA cm⁻² over 80 cycles and outstanding cycling stability. This work highlights the importance of A-site cations of the perovskite oxide as a valid strategy to improve ORR and OER activities.

1. Introduction

With the rapid consumption of fossil energy and the aggravation of environmental pollution, the development of efficient, green and sustainable energy storage and conversion technologies has attracted widespread attention due to the increasing demands for renewable and environmental-friendly energy. Particularly, rechargeable zinc–air batteries as a promising candidate for future energy application have attracted intensive attention due to the high safety, cost effective, high theoretical volumetric and gravimetric energy densities as well as environment-friendly [1,2]. Nevertheless, the performance improvement and commercialization of rechargeable zinc–air batteries were restricted by the sluggish kinetics of the oxygen reduction reaction (ORR) during the discharge process as well as the oxygen evolution reaction (OER) during the charge process at the air electrode [3,4].

Hitherto, noble-metal Pt catalyst has been regarded as the benchmark for ORR, and the oxides of Ru or Ir are regarded as the benchmark towards OER, which were greatly impeded by the exorbitant price, scarcity and unsatisfactory stability [5,6]. Furthermore, noble-metal-based catalysts usually exhibit selective activity for either ORR or OER, which poses a significant challenge to the ideal air electrode design [7]. Consequently, it is highly desirable to explore novel bifunctional catalysts for both ORR and OER with earth-abundant, environmentally friendly and high durability.

Growing attention has been given to perovskites with the general formula of “ABO₃” (A = alkaline-earth or rare-earth metals, B = transition metals) due to the high compositional flexibility, abundance on earth as well as the relatively high ionic and/or electronic conductivities [8,9,10]. The binding strength of reaction intermediates of ORR/OER, closely related to the intrinsic electronic structure of the catalyst, is the key factor affecting electrocatalytic activity. The superior catalytic activity could be achieved when the adsorbed species lightly bind to the surface neither too strongly nor too weakly according to the Sabatier’s principle [11]. Therefore, electronic structure descriptors were applied to correlate the ORR and/or OER, such as the O 2p band center, the charge-transfer energy and the e_g-orbital filling [12,13,14,15]. The e_g-orbital filling could be the most suitable descriptor to describe ORR and OER activities of perovskite oxides according to the molecular orbital theory considerations and experimental observations [16]. A volcano-type relationship between the e_g-orbital filling and catalytic activity of transition metals was reported by Suntivich et al. [14]. In addition, the e_g-orbital filling of the catalyst with highest activities was usually close to 1.2. The binding energy of *OH− is too strong when the e_g-orbital filling is higher than 1, and too weak when the eg-orbital filling is lower than 1. Perovskites allows the isovalent and aliovalent doping to adjust the e_g-orbital filling due to the high Goldschmidt tolerance factor, thereby regulating ORR/OER catalytic activities [15,17]. Many efforts have been devoted to improving the bifunctional catalytic activity and cycling stability of perovskite oxides by means of doping in A-site and/or B-site cations [18,19,20]. LaMnO₃, LaCoO₃, La_0.6Sr_0.4Co_0.4Fe_0.6O₃ and La_0.5Sr_0.5CoO₃ perovskite oxides have been evaluated as bifunctional catalysts, and the results demonstrated that the change of B-site cations would lead to the change of the oxygen adsorption energy on the surface of bifunctional catalysts, thus affecting catalytic activities [21,22]. Deficiency and/or doping of A-site cations could regulate the oxygen vacancy concentration, the mixed ionic and electronic conductivity and the valence of B-site cations, which is closely related to catalytic activities of ORR and OER [23,24]. Perovskites with moderate A-site-deficiency and/or doping, such as La_1-xNiO₃, La_1-xFeO₃, La_1-xCoO₃, and La_1-xSr_xCoO₃, displayed enhanced oxygen redox catalytic activities due to the mixed-valence states of B-site cations, the mixed ionic and electronic conductivity, the enhanced σ* electron-transferring ability as well as the hydroxide/oxygen surface-adsorbing ability [25,26,27,28]. However, excessive A-site defects in perovskite will lead to the formation of the second phase, which is unfavorable to bifunctional catalytic activities.

Several previous reports have shown that the A-site cations have a great influence on properties of perovskite oxides such as LnBaCo₂O_5+δ (Ln = La, Pr, Nd and Sm), Ln_0.6Sr_0.4CoO_3-δ (Ln = La, Pr, Nd, Sm and Gd) and Ln_0.5Sr_0.5Fe_0.9Mo_0.1O_3-δ (Ln = La, Pr, Nd, Sm and Gd) [29,30,31]. The electronic and ionic conductivities, oxygen vacancy concentration and the electrocatalytic activity of perovskite oxides were affected by A-site cations [27,31]. Therefore, it is necessary to systematically explore the influence of various A-site cations on ORR and OER catalytic activities of perovskite oxides.

Given the foregoing, the effect of A-site rare earth cations on the crystal structure, the valence state of Co and bifunctional catalytic activities of Ln_0.5Sr_0.5CoO_3-δ (Ln = La, Pr and Sm) prepared by the facial sol-gel method were systematically investigated in this study. The phase structure and compositions transformation of Ln_0.5Sr_0.5CoO_3-δ were analyzed in detail. The electrocatalytic properties of the as-prepared Ln_0.5Sr_0.5CoO_3-δ for both ORR and OER were systematically tested in the alkaline medium. Furthermore, electrocatalytic properties and the stability of as-prepared catalysts were also evaluated in home-made rechargeable zinc–air batteries.

2. Results and Discussion

The XRD measurement was applied to reveal the crystal structure of as-synthesized samples. Room temperature XRD patterns of La_0.5Sr_0.5CoO_3-δ (LSC) and Pr_0.5Sr_0.5CoO_3-δ (PSC) were displayed in Figure S1a,b, which agreed well with the corresponding standard cards (PDF#48-0122, PDF#53-0110), without any impurity peaks, indicating that the pure powders were synthesized successfully. The XRD pattern of Sm_0.5Sr_0.5CoO_3-δ (SSC) (Figure S1c) displayed that the main diffraction lines coincide with the corresponding standard card (PDF#53-0112). However, the non-indexed peak labeled with an asterisk at 2θ of 33.87° was related to a small amount of SmCoO₃ (PDF#25-1071) [32]. To reveal the characterization clearly, XRD patterns and the enlarged view at the main peak with 2θ = 32–35° of LSC, PSC and SSC (Figure 1). Importantly, some peaks shift positive in the order of LSC, PSC and SSC. It is clearly to be seen that the diffraction peaks gradually shift to higher angles in the order of LSC, PSC and SSC. The crystal structure distortions increase as the lanthanide ionic radii decreases from La to Sm. LSC displayed only a single peak in the (110) direction. Similarly, PSC exhibited only a single peak in the (112) direction. However, the peak of SSC split into two peaks, namely the (200), (121) and (002) directions, which belongs to phase transition from tetragonal to orthorhombic, indicating the distortion of crystal structure toward lower symmetry as previous similar results [29]. Perovskite-type structure with cubic symmetry is regarded as ideal perovskite structure with the highest phase stability. The structural symmetry of ABO₃ perovskite can be evaluated by the Goldschmidt tolerance factor t defined as follows Equation (1) [33]:

$$t=\frac{r_A+r_O}{\sqrt{2}(r_B+r_O)}$$

(1)

where $r_A$, $r_O$, and $r_B$ are the mean ionic radius of Ln³⁺ and Sr²⁺, oxygen ionic radius and the ionic radius of cobalt, respectively. The ionic radii of La³⁺, Pr³⁺ and Sm³⁺ are 1.36, 1.30 and 1.24 Å, respectively [27,34]. With a smaller ion as the A-site cation, i.e., for Pr³⁺ and Sm³⁺, the structure of perovskite oxides deviated from cubic and was closer to orthorhombic [27]. The gradual decrease of t value caused by the decrease of Ln³⁺ ionic radius from La³⁺ to Sm³⁺, suggesting the lower crystal symmetry and the decrease of the cell volume [31].

The morphologies of the as-prepared LSC, PSC and SSC powders were presented in Figure S2 and Figure 2. It is clearly seen that LSC, PSC and SSC powders with the size of about several hundred nanometers were agglomerated into larger particles of several micrometers with porous structure due to the high sintering temperature. It should be noted that PSC exhibits a porous structure, which is conductive to the adequate exposure of active sites during ORR and OER.

The cation oxidation state and the surface chemical composition of LSC, PSC and SSC were analyzed by XPS [35]. The carbon peak located at 284.71 eV could be the residual carbon after calcination of as-prepared samples [15]. The survey spectrum of LSC displayed peaks of La 3d, Sr 3d, Co 2p, O 1s and C 1s peaks according to XPS results in Figure S3a. Similarly, the survey spectrum of PSC and SSC displayed peaks of Pr 3d and Sm 3d, respectively. The corresponding binding energies of various peaks were consistent with previously reported values [36,37].

The oxygen-related species as well as the transition metals on the surface of the perovskite oxide were usually regarded as active sites to ORR and OER [38]. Two main peaks of O 1s spectra for LSC, PSC and SSC indicated that more than one type of oxygen species presented on surface of various catalysts as shown in Figure 3a–c.

Although in the O 1s spectra the positions of peaks of various catalysts remain unchanged, the relative intensity of peaks has obviously changed. The O 1s spectra of LSC, PSC and SSC were deconvoluted into four peaks. The peak located at the lower binding energy around 528.25 eV was ascribed to the lattice oxygen species O²⁻ in LSC, PSC and SSC (Figure 3a–c). The relative intensity of the lattice oxygen species O²⁻ for PSC was higher than that of LSC and SSC. Previous studies have suggested that the stronger covalency of lattice oxygen species was helpful to improve the ORR kinetics [39]. Taking PSC as an example, the higher binding energy peak for PSC was deconvoluted into three peaks, which ascribed to characteristics of O₂²⁻/O⁻ (529.99 eV), hydroxyl groups OH⁻/O₂ (531.18 eV) and the molecular water (532.35 eV), respectively [40,41]. Generally, the ${\mathrm{O}}_2^{2-}/{\mathrm{O}}^{-}$ was regarded as reactive species for ORR and OER [42].

The best OER activity was predicted to be at the e_g-orbital filling close to 1, which could promote the binding of oxygen-related intermediate species on B-sites and thus improve OER activities [14]. The e_g-orbital filling of Co³⁺ ($t_{2g}^5{e}_g^1$) and Co²⁺ ($t_{2g}^5{e}_g^2$) was 1 and 2, respectively. The surface electronic structure is affected by the oxidation state of the surface transition metal cations [36]. Hence, a high Co³⁺/Co²⁺ ratio can achieve a near-unity of e_g-orbital filling [43,44]. Higher oxidation state of Co in the perovskite oxide is regarded as potential sites for oxygen redox reactions [45,46]. The Co 2p spectrum of LSC, PSC and SSC can be deconvoluted into major characteristic peaks and two shake-up satellites (Figure 3d–f). For PSC, the peaks located at 779.50 eV and 794.53 eV are assigned t°Co³⁺, and the peaks at 780.69 eV and 795.76 eV indicate the existence of Co²⁺ as shown in Figure 3e. The area under the XPS peaks was fitted, which corresponds to their relative amounts. Among the as-prepared catalysts, displayed the Co³⁺/Co²⁺ ratio of PSC was 1.36 as shown in Table S1, which was much higher than that of LSC (1.02) and SSC (0.98). The higher Co³⁺/Co²⁺ ratio was conductive to the promotion of the electron transfer process during ORR and OER, which led to the enhancement of ORR and OER catalytic activities [35,36,47].

LSV curves were tested by rotating disk electrode in O₂-saturated 0.1 M KOH to evaluate catalytic activities of LSC, PSC and SSC for ORR. Cyclic voltammetry curves of LSC, PSC and SSC in 0.1 M KOH saturated with O₂ were displayed in Figure 4a. It can be seen that PSC displayed more positive oxygen reduction peak potential (0.58 V vs. RHE) than that of the LSC (0.57 V vs. RHE) and SSC (0.56 V vs. RHE). ORR activities of various samples were evaluated by LSV curves at a rotating rate of 1600 rpm as shown in Figure 4b. PSC displayed the half-wave potentials (E_1/2) about 0.55 V vs. RHE. Furthermore, the limiting current density of various samples follows the trend of SSC (3.34 mA cm⁻²) < LSC (4.04 mA cm⁻²) < PSC (4.19 mA cm⁻²) at 0.40 V vs. RHE. Therefore, PSC displayed significantly enhanced ORR catalytic activity than that of LSC and SSC. LSV curves were measured at various rotating rates as shown in Figure 4c to further explain the mechanism of PSC for ORR. The electron transferred numbers (n) calculated by K-L plots, as shown in Figure 4d, was about 3.94, which indicated the good selectivity for four electron ORR pathways.

In addition to the outstanding ORR activity discussed above, the OER activity of various samples was also evaluated in O₂-saturated 0.1 M KOH using a 10 mV s⁻¹ scan rate and 1600 rpm rotation rate. The corresponding polarization plots were iR-corrected to compensate for the solution resistance and depicted in Figure 5a. It can be seen that PSC displayed satisfactory OER activities with the potential of 1.75 V vs. RHE at the current density of 10 mA cm⁻², which was lower than that of LSC (1.77 V vs. RHE) and SSC (1.79 V vs. RHE). Therefore, PSC exhibited better OER performance than LSC and SSC. Moreover, the OER electrocatalytic kinetics of various bifunctional electrocatalytic were further investigated by the Tafel slope as shown in Figure 5b. PSC displayed the Tafel slope of 105 mV dec⁻¹, which was smaller than that of LSC (125 mV dec⁻¹) and SSC (120 mV dec⁻¹), indicating its superior OER activities. The potential gap (ΔE = E_OER@10 − E_1/2) between the OER potential at 10 mA cm⁻² (E_OER@10) and the ORR half-wave potential was used to evaluate the electrochemical reversibility of the oxygen-related electrode [48,49]. The ΔE value of PSC was 1.20 V (Figure 5c), which implied its reversibility and better bifunctional activity [50].

Home-made zinc–air batteries were fabricated to assess the bifunctional activity and stability of as-prepared catalysts under actual operating conditions (Figure S4). All home-made zinc–air batteries were conducted in ambient conditions (room temperature) using atmospheric air. The primary zinc–air battery with PSC demonstrated a stable open circuit voltage of 1.47 V as illustrated in Figure 6a, indicating PSC as promising candidate in practical implementation [51]. As shown in Figure 6b, the primary zinc–air battery with PSC displayed the maximum power density of 72 mW cm⁻² at the current density of 120 mA cm⁻², which was higher than that of battery with SSC (54 mW cm⁻² at the current density of 92 mA cm⁻²) and LSC (57 mW cm⁻² at the current density of 93 mA cm⁻²).

In addition to the galvanodynamic behavior, the kinetics of the as-prepared air electrodes were evaluated by the electrochemical impedance spectroscopy as shown in Figure 6c. Nyquist plots of various samples for primary zinc–air batteries at 1.2 V vs. Zn consists of two semicircles, which is similar to the previous reports [52,53]. Nyquist plots were fitted by an equivalent circuit (Figure 6c, inset) with five elements (R_ohm, R_int, R_ct, CPE_int, CPE_ct) [53,54,55]. R_ohm represents the electrolyte and contact resistance of primary zinc–air batteries. R_int represents the solid–liquid electrolyte interface resistance, which was related to the electrical conductivity of catalyst as well as its wettability in the liquid electrolyte [56]. In addition, the charge-transfer resistance (R_ct) was related to the catalytic activity of the catalyst during electrochemical reactions [56]. Constant phase elements (CPE_int and CPE_ct) were related to the capacitance at the interface of the air electrode and the liquid electrolyte [56]. On the basis of the equivalent circuit, the resistance values (R_ohm, R_int and R_ct) were listed in Table S2. The primary zinc–air battery with PSC displayed lower R_ohm (1.46 Ω) than that of the primary zinc–air battery with LSC (1.70 Ω) or SSC (1.64 Ω), which was mainly attribute to the higher electronic conductivity of PSC than LSC and SSC as previously reported [24,57]. PSC displayed the lowest charge-transfer resistance (0.929 Ω) than that of LSC and SSC, which indicated the improvement of ORR kinetics.

Similarly, primary zinc–air batteries with PSC as the air electrode displayed the discharge time about 23 h at the constant current density of 10 mA cm⁻² (Figure 6d). Being as a practical example, the LED lamp was successfully light up with two zinc–air connected in series assembled with PSC air electrode (inset of Figure 6d). The recharge ability and stability of various catalysts as the air electrodes were also evaluated by rechargeable zinc–air batteries in an atmospheric air environment. The overpotential for ORR and OER of zinc–air battery with PSC as bifunctional catalysts is 1.48 V at 50 mA cm⁻² as discerned in Figure 7a, which is significantly lower than that of SSC and LSC. The charge-discharge cycling stability measurements of rechargeable zinc–air batteries with LSC, PSC and SSC as air electrodes were performed at 10 mA cm⁻² for 30 min discharge and 30 min charge for 80 cycles (80 h). After 80 cycles, rechargeable zinc–air batteries with PSC as the air electrode exhibited the lowest charge-discharge voltage gap (1.01 V, Figure 7b), further indicating the promising bifunctional catalytic activities and excellent cycling stability.

3. Experimental

3.1. Material Synthesis

All reagents in this work were the analytical grade, which were used as received without further purification. Ln_0.5Sr_0.5CoO₃ (LnSC, Ln = La, Pr and Sm) were synthesized via the sol-gel method according to previously reported synthesis procedures [54]. Specifically, La(NO₃)·6H₂O (99.9%), Pr(NO₃)₃·6H₂O (99.9%), Sm(NO₃)₃·6H₂O (99.9%), Sr(NO₃)₂ (99.9%) and Co(NO₃)₂·H₂O (99.9%) were purchased from Innochem (Beijing, China). Citric acid (C₆H₈O₇·H₂O) and zinc acetate ((Zn(CH₃COO)₂·2H₂O, 99%)) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Typically, the nitrates of the various metals (La, Pr, Sm, Sr and Co) in the required stoichiometric concentration were dissolved in distilled water. Then, citric acid was added in the molar ratio of 1.5:1 to the metal ion. The solution was stirred continuously at 80 °C under the water bath for 4 h to obtain the viscous wet gel. After that, the solid precursor was obtained by drying the wet gel at 200 °C for 12 h. The resulting black powder was received by calcination of the solid precursor at 1100 °C for 4 h under the air atmosphere.

3.2. Physical Characterization

X-ray diffraction (XRD, Rigaku D/max2200, Tokyo, Japan) with Cu-Kα radiation (λ = 0.15418 nm) was applied to analyze the phase structure of LSC, PSC and SSC. The contents and valence state information were analyzed by the X-ray photoelectron spectroscopy (XPS, Thermo Scientific Escalab 250XI, Waltham, MA, USA), Al Kα radiation, hν = 1486.6 eV). The C 1s binding energy (284.8 eV) was applied to compensate the binding energy of all XPS peaks in this work. The morphology of various catalysts was evaluated by the transmission electron microscopy (TEM, Tecnai G2 20ST, Amsterdam, The Netherlands) and a field emission scanning electron microscope (SEM, Hitachi-SU800, Tokyo, Japan).

3.3. Electrochemical Measurements

All electrochemical measurements in this work were performed with the CHI 660E electrochemical workstation (Chenhua, Shanghai, China). The RRDE-3A Apparatus (ALS, Tokyo, Japan) was applied to test the electrochemical properties in the three-electrode half-cell consisting of the Hg/HgO reference electrode, a working electrode based on the rotating disk glassy carbon electrode (diameter: 5 mm, geometric area: 0.196 cm²) and the Pt wire counter electrode. 5 mg of catalyst as well as 5 mg of conductive carbon (Super P Li, TIMCAL) were ultrasonically dispersed into the mixture of 50 μL of Nafion solution (5 wt %) and 500 μL of ethanol and sonicated for at least 1 h to form a uniformly dispersed ink as previously reported [54,58]. Subsequently, 5 μL of as-prepared catalysts catalyst ink was dropped onto the surface of glassy carbon disk electrode. The nominal catalyst loading of the as-prepared catalyst was 232 μg cm⁻². All electrochemical properties toward ORR and OER of as-prepared catalysts were evaluated by the linear sweep voltammetry (LSV) at various rotating rates in O₂-saturated 0.1 M KOH with the scan rate of 10 mV s⁻¹.

The results of potentials were given versus reversible hydrogen electrode (RHE) according to the following Equation (2) [59]:

E(vs. RHE) = E(vs. Hg / HgO) + 0.098 + 0.059 × pH

(2)

The electron-transfer number(n) during ORR can be calculated based on the Koutechy–Levich (K-L) Equations (3)–(5) given below [60,61]:

$$\frac{1}{j}=\frac{1}{j_{\mathrm{L}}}+\frac{1}{j_{\mathrm{K}}}=\frac{1}{B}{\omega }^{-1/2}+\frac{1}{j_{\mathrm{K}}}$$

(3)

$$B=0.62nF{C}_0{(D_0)}^{2/3}{\nu }^{-1/6}$$

(4)

$$j_{\mathrm{K}}=nFk{C}_0$$

(5)

Here, j_L, j_K and j are the kinetic-limited, mass-transfer-limited and measured current densities, respectively. F is the Faraday constant (F = 96485 C mol⁻¹). k is the Boltzmann constant. n is the number of transferred electrons. C₀ is the saturated concentration of oxygen in electrolyte (C₀ = 1.26 × 10^-6 mol cm⁻³). ν is the kinetic viscosity of solution (ν = 0.01009 cm² s⁻¹), D₀ is the diffusion coefficient of oxygen. ω is the rotating rate (rad s⁻¹). The proportionality coefficient (B) is determined from the slope of K-L Equation. j_K is assumed to be the constant at the certain potential [62]. j_L is proportional to ω^−1/2. Hence, the electro-transfer number (n) can be calculated according to the slope of the linear relationship of j⁻¹ versus ω^−1/2.

The Tafel slope (b) was calculated depending on the Tafel curve derived from the LSVs of OER measurements based on the Tafel Equation (6) [63]:

$$\eta =a+b\times \log j$$

(6)

where a and j was the Tafel constant and measured current density, respectively. The overpotential (η) was obtained based on η = E_{vs. RHE} − 1.23.

3.4. Characterization of Rechargeable Zinc–Air Batteries

Home-made zinc–air batteries were assembled to characterize the bifunctional catalytic activities under ambient atmosphere. The home-made air electrode was composed of a catalyst layer, a current collector and a hydrophobic gas diffusion layer. The catalyst layer slurry was prepared by mixing the active carbon, the as-prepared catalyst powder, acetylene black and polytetrafluoroethylene emulsion (PTFE, 60 wt %) with the weight ratio of 3:3:1:3 into ethanol as per the previous report [64]. The gas diffusion layer slurry consisted of acetylene black and PTFE with the mass ratio of 2:5. The catalyst layer slurry and the gas diffusion slurry were continuously magnetically stirred in the water bath at 80 °C until the formation of a dough-like mixture, respectively. Afterwards, the gas diffusion layer slurry and catalyst layer slurry were evenly coated on each side of the foam nickel. The air electrode was dried at 80 °C for 12 h. The introduction of acetylene black as the composite partner is conducive to the formation of heterogeneous interface and the electronic transfer at the interface, which helps to improve the efficiency and stability of the air electrode [65,66]. The polished zinc plate with the thickness of 1 mm was applied as the anode of zinc–air batteries. The separator was not used in this work. The electrolyte of primary zinc–air batteries was 6 M KOH aqueous solution. However, the electrolyte of rechargeable zinc–air batteries was the solution containing 6 M KOH and 0.2 M zinc acetate. The polarization and power density curves as well as the electrochemical impedance spectroscopy of zinc–air batteries were recorded by the electrochemical workstation (DH7000, Jiangsu Donghua Analytical Instruments Co., Ltd., Taizhou, China). The discharge as well as charge-discharge cycling tests were recorded by the Neware Battery test instrument (CT-4008-5V6A-S1, Shenzhen Neware Instrument Company, Shenzhen, China).

4. Conclusions

In summary, LSC, PSC and SSC were fabricated by the so-gel method and systematically evaluated as bifunctional catalysts for ORR and OER. A-site cations of perovskite oxides not only causes the distortion of crystal structure, but also changes the valence state of Co. Compared with LSC and SSC, PSC demonstrated the enhanced catalytic activity towards ORR (the limiting current density of 4.19 mA cm⁻² at 0.40 V vs. RHE) and OER (the potential of 1.75 V vs. RHE at 10 mA cm⁻² and the Tafel slope of 105 mV dec⁻¹). The higher Co³⁺/Co²⁺ ratio of PSC was conductive to the promotion of the electron transfer process during ORR and OER, thereby improving ORR and OER catalytic activities. Rechargeable zinc–air batteries with PSC as the air electrode displayed the maximum power density of 72 mW cm⁻² and the low charge-discharge voltage gap (1.01 V) over 80 cycles, which mainly due to the improvement of charge-transfer process. This work not only unveils A-site cations effects on the bifunctional catalytic activities of perovskite oxides but also provides a valid strategy to explore novel bifunctional catalysts for rechargeable zinc air batteries.