Controllable loading of an Fe/Co alloy on heteroatom-doped hollow graphene spheres realized via regulation of small molecules for rechargeable zinc–air batteries

Abstract

Although the use of transition metals as bifunctional catalysts for zinc–air batteries (ZABs) has obvious economic advantages, their performance in ZABs still fails to meet expectations due to the uncontrollable loading caused by the rapid nucleation rate of transition metals. In this study, controllable loading of an Fe/Co alloy on heteroatom-doped hollow graphene spheres (FeCo@NGHS) was realized via the regulation of small molecules. Sodium citrate, which served as a metal complexing agent and reaction buffer, effectively suppressed the excessive loading of Fe/Co alloy particles and facilitated the formation of Fe(Co)N_x active sites. Melamine, which served as a precursor for doping N atoms, provided anchor points for the loading of Fe/Co alloy particles and participated in the generation of Fe(Co)N_x. The fabricated catalyst had active sites with different chemical structures, such as pyridine-N, graphite-N, Fe(Co)N_x and Fe/Co alloy particles, all of which benefit the improvement of the oxygen reduction reaction/oxygen evolution reaction (ORR/OER) performance. Results showed that the fabricated FeCo@NGHS, which possesses the appropriate amount of Fe/Co alloy particles combined with the highest amount of formed Fe(Co)N_x active sites, exhibited the best ORR/OER bifunctional catalytic performance in alkaline electrolytes and excellent electrocatalytic stability. The ORR onset potential and half-wave potential were 0.961 V and 0.846 V (vs. RHE), respectively. The OER could achieve a low overpotential level of 391 mV at a current density of 10 mA cm⁻². Furthermore, the rechargeable liquid ZAB and flexible all-solid-state (ASS) ZAB assembled by FeCo@NGHS exhibited higher discharge power density and longer charge–discharge cycle performance. FeCo@NGHS-based air cathodes exhibited outstanding performance in flexible ASS–ZABs, showing high open circuit voltage (1.45 V) and peak power density (74.06 mW cm⁻²). Thus, in clean energy storage and conversion technologies, a new synthetic strategy for constructing excellent bifunctional oxygen electrocatalysts is proposed in this work.

---

Introduction

In the current context of fossil fuels, such as coal, oil shale, petroleum and their processed products, as power sources, there are numerous environmental factors to consider for human society.^1–4 Zinc–air batteries (ZABs) with a simple preparation process are gaining interest as a promising energy storage and conversion device in meeting the growing energy demand. ZAB has been extensively studied owing to its desirable properties, such as cost-effectiveness, environmentally safe operation with no pollution output, and a high theoretical energy density.⁵ During the charge/discharge cycles, reactions at the air cathode encompass the oxygen reduction reaction/oxygen evolution reaction (ORR/OER), but the ZABs’ market entry is hindered by the sluggish electron transfer rates and substantial overpotentials for these critical reactions.^6,7 Hence, developing efficient electrochemical catalysts showcasing improved ORR and OER activities is crucial for the progress of ZABs. Presently, some noble metal electrocatalysts, including Pt-based and RuO₂/IrO₂-based formulations, demonstrate exceptional catalytic efficacy in ORR and OER, respectively.^8–11 However, the extremely high cost, lack of reserves and unsatisfactory stability of noble metals have gradually made them unattractive.^12–14 Therefore, it is important to explore bifunctional oxygen electrocatalysts, which are efficient, inexpensive and have excellent stability in ZABs.¹⁵

Transition metal-based materials have garnered attention in recent years for their impressive electrocatalytic performance and durability, making them promising candidates for various electrochemical applications.^16–20 The raw materials of transition metals (such as Fe and Co) are all abundant and inexpensive.^21,22 On the one hand, Fe-based catalysts can show excellent activity in ORR, while iron nitride-based carbon materials can even be comparable to the Pt-based catalysts.²³ On the other hand, Co-based catalysts are highly promising in the OER active catalysts, including Co metal, cobalt oxide, and cobalt phosphide,^24,25etc. Therefore, the combination of Fe and Co is expected to be an efficient and low-cost bifunctional catalyst. However, nanoparticles (NPs) of the above metals suffer from dissociation, migration, and aggregation problems, resulting in the decay of catalyst activity and stability. Researchers usually construct structural defects and introduce heteroatoms to anchor metal NPs on the surface of carbon-based materials, and also use pore-framing NPs as a way to achieve effective control of NPs.^26–28 In addition, the low conductivity of the transition metals-based materials also affects their catalytic performance. Considering the need for material conductivity in electrocatalysis, catalyst substrates are often chosen from carbon nanotubes, graphene, etc.^29–31 Except the transition metal, metal–N–C (metal = Fe, Co, Ni) has received much attention due to its high activity, which is centered on a metal–nitrogen mimicking a bioporphyrin center coordination structure. Researchers have reported on many excellent performing M–N–C electrocatalysts.^32–35 The group of Kurungot prepared an Fe/Co-rich nitrogen-doped active site (Fe-/Co-NpGr) nanoporous graphene for ORR catalysis.³⁶ Through a carefully controlled oxidative etching procedure, graphene was transformed into porous graphene (pGr). The resulting edge sites were used to dope nitrogen, and thus construct Fe/Co coordination centers (Fe/Co-NpGr) on the basis of nitrogen doping. X-ray photoelectron spectroscopy (XPS) and time-of-flight secondary mass spectrometry (ToF-SIMS) analyses revealed structural insights, confirming the existence of the M–N (M = Fe, Co) doped carbon active site ligands. The results show that the bimetallic ligand locked at the hole opening can significantly reduce the overpotential of the ORR during the reaction with the assistance of nitrogen.

Herein, we report a facile and environmentally friendly one-step method for the controllable loading of an Fe/Co alloy on heteroatom-doped hollow graphene spheres (FeCo@NGHS), which can be realized by small molecule regulation and effectively avoid the agglomeration of transition metal particles (as shown in Fig. 1). In this work, graphene hollow spheres were used as the carbon matrix due to the good conductivity, high specific surface area and excellent flexibility. Inspired by the synthesis process of Prussian blue analogues (PBA), sodium citrate was used as the metal complexing agent and reaction buffer for the controllable loading of Fe/Co alloy. Because the Co²⁺ ions can coordinate with sodium citrate, which is a competitor of [Fe(CN)₆]³⁻, to moderate the proliferation rates during the nucleation and subsequent growth phases. According to the documented research, the synthesis of Co–Fe PBA proceeds at a swift pace. Without the addition of sodium citrate, nucleation takes place immediately after the direct mixing of Co²⁺ and [Fe(CN)₆]³⁻, which generates a large number of nuclei and particles. The coordination of Co²⁺ ions with sodium citrate can slow the reaction kinetics, thus leading to the controllable loading of the Fe/Co alloy. Moreover, cobalt acetate and potassium ferricyanide work as the cobalt and iron sources, respectively. Nitrogen atoms can be introduced into the NGHS mainly with the help of melamine. The doping of nitrogen atoms with different chemical structure and the riveting of the Fe/Co alloy are closely related to the introduced sodium citrate and melamine. The designed catalyst FeCo@NGHS shows excellent ORR and OER performance, where the loaded Fe/Co alloy combined with heteroatom active sites, such as pyridine-N, graphite-N, and Fe(Co)N_x, all played an active role in the enhancement of the bifunctionality. The rechargeable liquid ZAB and flexible all-solid-state (ASS) ZAB assembled from FeCo@NGHS exhibit high discharge power density and more durable charge–discharge cycling performance. The FeCo@NGHS-based air cathode has outstanding performance in the flexible ASS–ZAB, presenting high open-circuit voltage (OCV) (1.45 V) and strong peak power density (74.06 mW cm⁻²).

Fig. 1 Conceptual diagram outlining the fabrication procedure of FeCo@NGHS.

Results and discussion

Characterization of the catalyst morphology and exploration of their formation mechanisms

First, the negatively charged graphene oxide (GO) and positively charged polystyrene spheres (PS) are mixed together with the presence of sodium citrate and melamine to form a homogeneous aqueous solution system. GO and PS are combined together through electrostatic force and form a GO@PS core–shell structure. The PS serve as a template for the formation of spherical hollow graphene, while melamine provides the nitrogen source for the doping of N atoms. During the electrostatic coating process, the PS could be inserted into the GO carbon flakes and effectively prevent the re-bonding of the GO sheets during the subsequent heat treatment, thus increasing the specific surface area of the resulting substrate material. Thereafter, cobalt acetate and potassium ferricyanide are gradually introduced into the system.

Due to the complexation of sodium citrate,³⁷ cobalt ions react with potassium ferricyanide slowly. After annealing, the sample FeCo@NGHS was fabricated. And the SEM image of FeCo@NGHS is shown in Fig. 2a. Furthermore, the uniform distribution of elements N, Fe and Co can be clearly seen from the EDS elemental mapping image of FeCo@NGHS (Fig. 2j). In order to explore the specific effect of sodium citrate, a controlled experiment was carried out. The preparation process was similar to that of sample FeCo@NGHS, but without the addition of sodium citrate, and the resulting sample was named as Fe&Co@NGHS. Fig. 2d shows an obvious particle aggregation in sample Fe&Co@NGHS compared with FeCo@NGHS. The total load of metal in sample Fe&Co@NGHS (7.30 wt%) is also higher than that in sample FeCo@NGHS (4.68 wt%) (Table S1†). It can be assumed that without the introduction of sodium citrate, more metal particles are anchored to the base material. The lack of complexation between sodium citrate and cobalt ions leads to the rapid complexation of cobalt ions with ferricyanide ions, thus forming more alloy particles.

Fig. 2 SEM, TEM and HRTEM images of (a–c) FeCo@NGHS, (d–f) Fe&Co@NGHS, and (g–i) FeCo@GHS. (j) EDS element mapping image of FeCo@NGHS. The inset of (b) is a SAED pattern.

Moreover, a comparative sample was designed and prepared to further understand the effect of melamine on the synthesized catalyst material. Sample FeCo@GHS was obtained by a preparation process similar to that of sample FeCo@NGHS without the introduction of melamine. Fig. 2g shows that without the introduction of melamine, the spherical structure in sample FeCo@GHS is less obvious than that in FeCo@NGHS and Fe&Co@NGHS, which is due to the lack of intercalation of melamine into GO sheets during the preparation process. The total load of metal (3.63 wt%) in sample FeCo@GHS is also lower than those of samples FeCo@NGHS and Fe&Co@NGHS (Table S1†). The inclusion of melamine is intimately linked to the incorporation of the alloy, as inferred from the analysis. Therefore, we can infer that melamine has a riveting effect on metal particles, and related studies have also suggested the anchoring effect of the nitrogen source urea on metal particles.³⁸ Due to the presence of melamine, samples Fe@NGHS and Co@NGHS maintain a good spherical structure, as shown in Fig. S1a and c.† However, the load of a single metal for sample Fe@NGHS and Co@NGHS is about 0.82 wt% and 0.87 wt%, respectively, which is significantly reduced compared with sample FeCo@NGHS. Therefore, it can be supposed that under the same preparation conditions, compared with the single metal, the selected base material NGHS is more favorable for alloy anchoring.

Structure and composition of catalysts

Fig. 3a shows the XRD diffraction spectra of the obtained catalyst materials. Sample FeCo@NGHS has distinct diffraction peaks at 45.1°, 65.6° and 83.2°, which can be respectively attributed to the (110), (200) and (211) crystal planes of the Co_0.7Fe_0.3 alloy (JCPDS # 48-1818),³⁹ so it can be inferred that the loaded metal particles are mainly alloys with the composition of Co_0.7Fe_0.3. In addition, as shown in Fig. 2c, the crystal plane spacing of NPs in FeCo@NGHS is about 0.1982 nm, and the d-value approximates the standard lattice stripe (0.2007 nm, JCPDS # 48-1818) of the (110) crystal plane of the Co_0.7Fe_0.3 alloy. Moreover, the diffraction rings in the SAED image (shown in the inset of Fig. 2b) correspond to the (110), (200) and (211) crystal planes of the Co_0.7Fe_0.3 alloy, respectively, which further confirms that the composition of the particles loaded in the sample FeCo@NGHS is undoubtedly alloy Co_0.7Fe_0.3. The samples Fe&Co@NGHS and FeCo@GHS also show an obvious distinct diffraction peak at 45.1°, which is attributed to the (110) crystal plane of the Co_0.7Fe_0.3 alloy (JCPDS # 48-1818).³⁹ As shown in the HRTEM images in Fig. 2f and i, the crystal plane spacings of the loaded NPs in Fe&Co@NGHS and FeCo@GHS are about 0.2007 and 0.1994 nm, respectively, corresponding to the (110) crystal plane of the Co_0.7Fe_0.3 alloy, which further confirms that the component of supported particles is alloy Co_0.7Fe_0.3. From Table S1,† it can also be seen that the metal loading of samples Fe@NGHS and Co@NGHS is low. Therefore, no significant characteristic peaks corresponding to Fe/Co are observed in the XRD spectra (Fig. 3a).

Fig. 3 (a) XRD patterns, (b) TGA curves, (c) Raman spectra, and (d) N₂ adsorption/desorption isotherm curves, and the inset is the pore size distribution.

A distinctive peak at 2θ = 22°–30° is consistently observed across all samples, which is attributed to the (002) lattice plane of graphite carbon (JCPDS # 41-1487) and stems from the carbonaceous substrates NGHS or GHS.⁴⁰ High-resolution TEM (HRTEM) images in Fig. 2c, f, and i distinctly reveal lattice fringes consistent with the (002) plane of graphite. Different from other samples, Fe@NGHS has three diffraction peaks corresponding to graphite. The diffraction peak positioned at approximately 44° is attributed to the (100) plane of graphite carbon (JCPDS # 41-1487). Furthermore, an additional diffraction peak at approximately 15° is associated with the (001) plane of graphene carbon, indicating an interlayer spacing of 0.59 nm, a value derived from the Bragg's equation. The extended layer spacing compared with graphite results from the intercalation of the substance in the graphene oxide carbon sheets, such as PS spheres, melamine, sodium citrate and transition metal salts. Diffraction peaks at different positions indicate different carbon layer spacing, suggesting that the arrangement of carbon sheets in sample Fe@NGHS is the most disordered.

Fig. 3b shows the thermogravimetric analysis (TGA) of samples FeCo@NGHS, Fe&Co@NGHS and FeCo@GHS. Before the temperature reaches 350 °C, the mass percentage of the fabricated samples shows no significant decline, indicating the good thermal stability. When the temperature increases to 350–420 °C, significant material decomposition is found, which is due to the loss of the carbon substrate material at high temperature. After calcination in air at 800 °C, sample Fe&Co@NGHS shows the highest mass residue, suggesting the highest metal loading. In order to accurately measure the amount of loaded metal in the resulting samples, inductively coupled plasma mass spectrometry (ICP-MS) detection is adopted, and the result is listed in Table S1.† The total amount of loaded Fe/Co alloy in sample Fe&Co@NGHS is the highest, followed by sample FeCo@NGHS, and sample FeCo@GHS exhibits the lowest. The ICP-MS test results are consistent with the TGA results. It is interesting that the introduction of melamine is directly related to the amount of Fe/Co alloy. Another more important problem is that under the same conditions, the loading of the Fe/Co alloy is easier to achieve than that of a single metal, such as Fe or Co.

The Raman spectra of the resultant samples are shown in Fig. 3c, and all samples present the D and G bands at 1355 and 1584 cm⁻¹, respectively.⁴¹ The D and G bands are symbols of the degree of disordered and graphitization structure of the material, where the D peak intensity is related to the defect site (sp³-hybridized carbon), while the G peak intensity is related to the amount of sp²-hybridized carbon in the carbon matrix.³⁸ It is common to consider I_D/I_G as an evaluation parameter for the degree of graphitization of carbon materials.⁴² It is not difficult to find that compared with NGHS, the introduction of Co or the Fe/Co alloy significantly reduces the value of I_D/I_G for the resulting samples (such as FeCo@NGHS, Co@NGHS, Fe&Co@NGHS and FeCo@GHS), except for sample Fe@NGHS. The I_D/I_G value in sample Fe@NGHS increased significantly, consistent with the XRD result. It can be concluded that the introduction of Fe has a significant impact on the substrate material NGHS. A smaller value of I_D/I_G indicates that the carbon sheets are arranged in a more ordered manner, which signifies a heightened level of graphitization, resulting in improved electrical conductivity. At the same time, a higher value of I_D/I_G suggests the presence of more defect sites in the prepared sample.

Fig. 3d depicts the N₂ sorption/desorption isotherms and the corresponding pore size distribution profiles for the prepared samples. Compared with NGHS, the specific surface area of the obtained catalyst material is not significantly reduced after the metal loading. On the contrary, some of them even improved significantly, which should be caused by the insertion of sodium citrate and transition metal salts. Especially, FeCo@NGHS presents the highest specific surface area (617.4 m² g⁻¹), indicating the exposure of more active sites. As shown in Fig. 3d, all samples show a distinct type IV isotherm in the medium pressure region (P/P₀ = 0.5–1.0) with a sharp increase in nitrogen uptake at relatively low pressures,⁴³ which reveal the presence of both microporous and mesoporous structures in the material. These findings are in line with the pore size distribution data presented in the inset. The high specific surface area and rich micro/mesoporous structure of the fabricated catalyst can provide higher mass transport speed for the electrocatalytic reaction, while the microporosity accelerates the electron transfer,⁴⁴ which effectively enhances the electrocatalytic activity with the synergy of the high specific surface and micro/mesopores.

The XPS wide-scan full spectrum in Fig. 4a shows the elemental composition on the surface of the resulting catalyst materials. The high-resolution C 1s XPS spectra are shown in Fig. 4b. Each spectral line can be resolved into four distinct peaks, corresponding to C–C at 284.8 eV, C=N and C–O at 285.85 eV, C=O and C–N at 288.75 eV, and O–C=O at 292.4 eV, respectively.⁴⁵ Furthermore, the sp² hybridized C–C is the main component, which can suggest the graphite structure of substrate NGHS and indicate the good conductivity. Fig. 4c directly displays the successful doping of N atoms into the carbon skeleton.⁴⁶ Melamine serves as the main nitrogen precursor for the doping of N atoms. The doped N atoms, which are strongly electronegative, can cause charge rearrangement and result in more active sites. In the high-resolution N 1s XPS spectrum of FeCo@NGHS (Fig. 4c), it can be deconvoluted into five peaks at 398.45 eV, 399.38 eV, 400.14 eV, 400.98 eV, and 402.98 eV, which are attributed to pyridine-N (398.45 eV), Fe–/Co–N (399.38 eV), pyrrole-N (400.14 eV), graphite-N (400.98 eV), and oxidized-N (402.98 eV), respectively.⁴⁷ Furthermore, all control samples, except for NGHS and FeCo@GHS, contained a metal–N active site.⁴⁸ Additionally, as can be observed from Table S2,† the sample FeCo@NGHS exhibits the highest content of the Fe–/Co–N component.

Fig. 4 XPS spectra of FeCo@NGHS, Fe@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS and NGHS. (a) XPS wide scan full spectrum; High-resolution XPS spectra of (b) C 1s, (c) N 1s, (d) O 1s, (e) Fe 2p and (f) Co 2p.

According to Fig. 4a, all samples exhibit high nitrogen doping, except the sample FeCo@GHS, which is synthesized without the introduction of melamine. From Fig. 4c, it can be found that pyridine-N and graphite-N are the main structures of nitrogen doping in the obtained samples, and they can promote the positive shift of the ORR onset potential and increase the diffusion-limited current density, respectively.^49,50 Metal nitrides (Fe(Co)N_x) also show positive correlation properties with electrocatalytic performance. As shown in Fig. 4c and d, the sample FeCo@NGHS contains higher metal nitrides (Fe(Co)N_x) and adsorbed oxygen content than other control samples, which can promote the effect of the bifunctional catalysis (Table S3†).

The introduction of melamine is directly related to the doping of N, and the doping of N atoms can not only introduce the active site of N atoms, but also provide anchor points for metal loading. Comparing the metal loading in FeCo@GHS and FeCo@NGHS, it can be seen that the absence of melamine leads to a significant decrease in alloy loading (Table S1†). In addition, compared to FeCo@NGHS, the lack of sodium citrate in Fe&Co@NGHS leads to rapid nucleation and excessive loading of the alloy. Although the loading of the alloy in Fe&Co@NGHS is high, the amount of Fe(Co)N_x obtained is obviously lower than FeCo@NGHS. Therefore, it can be inferred that the complexation of sodium citrate with cobalt ions slows down the formation of alloy particles, thus promoting the generation of more Fe(Co)N_x active sites. The FeCo@GHS sample has fewer N and Fe(Co)N_x active sites compared to FeCo@NGHS due to the absence of melamine during its synthesis process, which is the main reason for its poor electrocatalytic performance.

As discussed above, the introduction of sodium citrate can achieve the controllable loading of alloy particles, and the riveting of the metal by doped N resulted in a moderate amount of metal loading in sample FeCo@NGHS, accompanied by the formation of a large number of Fe(Co)N_x active sites. Combining the above analytical results, it is easy to see that the introduction of sodium citrate and melamine is the key to form more Fe(Co)N_x active sites. It is interesting that metal–N_x active sites, such as Fe–N_x and Co–N_x types, have been confirmed by the previously reported research work for their strong coupling effect between N-doped graphene shells and metal NPs, accelerating the mass transfer efficiency and thus enhancing the catalytic performance.⁵¹ Therefore it can be assumed that the formation of Fe(Co)N_x is related to both melamine and sodium citrate, and the absence of either will reduce the N content and the percentage of Fe(Co)N_x. From Fe@NGHS, it can be seen that the doping amount of N is relatively high, but the loading amount of Fe is low, indicating that the loading of a single metal is difficult under this synthesis condition.

As shown in the high-resolution XPS spectra of Fe 2p (Fig. 4e), five pairs of double peaks are present in sample FeCo@NGHS, with zero-valence Fe peaks observed at 707.73 eV (Fe 2p_3/2) and 719.56 eV (Fe 2p_1/2). This indicates the presence of metallic Fe, which can be attributed to metallic Fe in alloy Co_0.7Fe_0.3. A pair of main peaks is located at 711.8 eV (Fe 2p_3/2) and 724.29 eV (Fe 2p_1/2), while satellite peaks are observed at 716.4 eV (Fe 2p_3/2) and 728.13 eV (Fe 2p_1/2), originating from Fe^2+/3+.³⁹ The Fe 2p high-resolution XPS spectra of samples Fe@NGHS and Fe&Co@NGHS also contain five pairs of double peaks, but there is no double peak representing Fe(Co)N_x in sample FeCo@GHS. Similarly, the high-resolution XPS spectra of Co 2p (Fig. 4f) show the presence of zero-valence Co peaks at 777.1 eV (Co 2p_3/2) and 792.9 eV (Co 2p_1/2), attributable to metallic Co in the alloy Co_0.7Fe_0.3. Co–N_x peaks are located at 778.4 eV (Co 2p_3/2) and 794.3 eV (Co 2p_1/2), while the satellite peaks are located at 786.0 eV (Co 2p_3/2) and 800.5 eV (Co 2p_1/2), respectively, suggesting the presence of the Co–N_x active sites.³⁹ Similarly, the Co 2p high-resolution XPS spectra of samples Co@NGHS and Fe&Co@NGHS include the same Co component, but the FeCo@GHS sample lacks the Co–N_x component. The high-resolution XPS spectra of the obtained sample FeCo@NGHS once again demonstrate the presence of Fe(Co)N_x active sites, as well as the metal and alloy composition in the loaded particles.

Electrocatalytic performance for ORR

According to the testing method of the three-electrode system, samples coated on rotating disk electrode (RDE) and rotating ring-disk electrode (RRDE) could be made into working electrodes, and tested for ORR performance in an O₂ saturated 0.1 M KOH electrolyte. The catalysts were first tested for LSV polarization curves, and the working electrode was maintained at 1600 rpm. As shown in Fig. 5a, compared with Fe@NGHS (E_onset = 0.91 V, E_1/2 = 0.769 V vs. RHE), Co@NGHS (E_onset = 0.885 V, E_1/2 = 0.81 V vs. RHE), Fe&Co@NGHS (E_onset = 0.962 V, E_1/2 = 0.83 V vs. RHE), FeCo@GHS (E_onset = 0.894 V, E_1/2 = 0.798 V vs. RHE), NGHS (E_onset = 0.842 V, E_1/2 = 0.739 V vs. RHE), and commercial Pt/C (E_onset = 0.932 V, E_1/2 = 0.829 V vs. RHE), FeCo@NGHS exhibited more positive E_onset (0.961 V vs. RHE) and E_1/2 (0.846 V vs. RHE), combined with the largest limiting current density. It showed the best ORR performance, which is also better than most reported transition metal-based carbon electrocatalysts (Table S4†).

Fig. 5 ORR performance of various catalysts, including FeCo@NGHS, Fe@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS, NGHS, and 20 wt% Pt/C in an O₂-saturated 0.1 M KOH solution. (a) ORR polarization curves; (b) EIS Nyquist plots; (c) Tafel plots; (d) CV curves (solid line: O₂-saturated, dashed line: N₂-saturated); (e) RRDE polarization curves; (f) electron transfer numbers and peroxide yields from RRDE measurements; (g) stability analysis; (h) combined ORR and OER polarization curves for FeCo@NGHS and Pt/C + RuO₂; (i) methanol and CO tolerance assessments for FeCo@NGHS and 20 wt% Pt/C (arrows indicate methanol and CO additions).

As shown in Fig. 5b, the EIS analysis of the samples FeCo@NGHS, Fe@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS, NGHS, and 20 wt% Pt/C were performed. It can be observed that the corresponding semicircle diameters of FeCo@NGHS, Fe@NGHS, 20 wt% Pt/C, Co@NGHS, Fe&Co@NGHS, FeCo@GHS and NGHS gradually increase, showing an increase in the charge transfer resistance (R_ct). FeCo@NGHS shows the smallest R_ct of 39.96 Ω (Table S5†), exhibiting the best ORR activity.⁵² Similarly, Fig. 5c shows that the Tafel slope of FeCo@NGHS is 86.56 mV dec⁻¹, which is lower than that of Fe@NGHS (157.8 mV dec⁻¹), Fe&Co@NGHS (104.29 mV dec⁻¹), FeCo@GHS (118.23 mV dec⁻¹), NGHS (122.23 mV dec⁻¹) and 20 wt% Pt/C (86.81 mV dec⁻¹), followed only by 65.11 mV dec⁻¹ from Co@NGHS. The lower Tafel slope indicates that the catalyst FeCo@NGHS has a reaction mechanism and kinetics similar to those of commercial Pt/C.⁵³ Compared with other bimetallic loaded samples, it can be assumed that more active sites of metal–N_x–C are formed in it due to the highest total N doping and percentage of Fe(Co)N_x in sample FeCo@NGHS. Therefore, the lower Tafel slope of FeCo@NGHS may by caused by the synergistic interaction between metal NPs and nitrogenous material, which accelerates the dissociation of OOH* (OOH* → O* + OH*), further highlighting the structural advantage of M–N–C.^54,55

From the CV curves in Fig. 5d, each catalyst exhibits a clear redox peak. The position of the oxygen reduction peak of the FeCo@NGHS (0.867 V) is higher than the potential of the rest of the control samples, such as Fe@NGHS (0.821 V), Co@NGHS (0.830 V), Fe&Co@NGHS (0.847 V), FeCo@GHS (0.810 V) and NGHS (0.741 V), which are consistent with the LSV comparison results. In the N₂-saturated electrolyte, none of the CV curves of the samples have ORR peaks, indicating that none of them had Faraday reactions. This also further confirms that the sample FeCo@NGHS has outstanding ORR performance.

As mentioned in the introduction, the catalytic activity of ORR catalysts is usually measured by the number of electron transfers, so the samples were tested for RRDE (the results are shown in Fig. 5e). The obtained data were converted to obtain Fig. 5f, where the electron transfer numbers and hydrogen peroxide yields of FeCo@NGHS, Fe@NGHS and Fe&Co@NGHS are close to those of Pt/C, indicating that the electron transfer mechanism of the catalysts is consistent with the four-electron pathway of commercial catalysts. The K–L plots in Fig. S2† present the linear correlation of the designed electrode catalysts at different potentials, indicating that the electrode catalysts have the primary reaction kinetics.⁵³

Through the comparison of the ORR performance, the ORR comprehensive performance order of the designed catalysts is as follows: FeCo@NGHS > Pt/C ≈ Co@NGHS > Fe&Co@NGHS > Fe@NGHS > FeCo@GHS > NGHS. FeCo@NGHS exhibits better performance than commercial Pt/C in terms of R_ct, kinetic current density, etc., which indicates that FeCo@NGHS has the best kinetic performance. On the one hand, the sample FeCo@NGHS has a moderate alloy loading due to the riveting effect of the doped N atoms. With the introduction of sodium citrate, the alloying reaction of potassium ferricyanide with cobalt proceeded controllably under the complexation of cobalt ions by sodium citrate, resulting in the formation of more Fe(Co)N_x active sites in FeCo@NGHS, compared with Fe&Co@NGHS. On the other hand, the FeCo@GHS without melamine incorporation exhibits poor performance, similar to the substrate material NGHS. This is mainly due to the lack of N doping, resulting in the decrease of the alloy load and the absence of a sufficient number of effective Fe(Co)N_x active sites. Moreover, the ORR activities of the single-metal samples Fe@NGHS and Co@NGHS are both weaker than those of FeCo@NGHS catalysts, due to the synergistic effect of the Fe/Co alloy, which can promote the ORR catalysis.

Durability is also critical for ORR catalysts. Fig. 5g shows the chronoamperometry response tests of FeCo@NGHS and 20 wt% Pt/C at 0.85 V. It can be observed that the catalyst FeCo@NGHS has good durability. Its relative current can still be maintained at 80.76% after 2.7 h of testing, while the commercial Pt/C is maintained at 76.32%. The methanol oxidation resistance and CO toxicity resistance of the catalysts were also measured. As shown in Fig. 5i, the current density of FeCo@NGHS or 20 wt% Pt/C decreased after adding methanol or passing CO. However, FeCo@NGHS did not change drastically compared with the noble metal catalysts, confirming that it has better resistance to methanol oxidation and the anti-CO toxicity ability. It further illustrates its application potential in fuel cell systems. The good activity and durability of the catalyst originate from the protection by the graphite structure and carbon matrix. After continuous ORR operation, the graphite substrate of FeCo@NGHS remained unchanged. Fig. S3† shows the SEM and TEM results, where the complete graphene sphere morphology could be observed. Metal NPs are still dispersed and supported in NGHS without obvious aggregation. As shown in the SAED (inset of Fig. S3b†), the diffraction rings corresponding to the crystal planes (211), (200) and (110) of the Fe/Co alloy still exist. After the ORR stability test, the XPS evaluation of the sample FeCo@NGHS was conducted to analyze its surface element composition and electronic structure. It can be observed from Fig. S4† that the overall elemental composition of the FeCo@NGHS remains unchanged, and the electronic structures of Fe and Co elements on the surface also remain unchanged.

Electrocatalytic performance for OER

To evaluate the OER performance of FeCo@NGHS, the catalyst was tested in 0.1 M KOH at a scan rate of 5 mV s⁻¹. As shown in Fig. 6a, for FeCo@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS and the NGHS catalysts, their presented OER overpotentials (η₁₀) are 391, 487, 452, 517 and 578 mV when reaching a current density of 10 mA cm⁻². Meanwhile, the Fe@NGHS sample shows almost no OER activity. From these data, it can be found that the catalyst FeCo@NGHS exhibits the best OER performance, which is on par with commercial RuO₂ (411 mV) and a multitude of reported transition metal-based carbon electrocatalysts, as detailed in Table S4.† EIS analysis was performed on the interfacial charge transfer kinetics, as shown in Fig. 6b. From the EIS Nyquist curves, the R_ct of FeCo@NGHS is lower than those of Fe@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS, NGHS and RuO₂ (Table S6†), indicating that FeCo@NGHS can accelerate the charge transfer kinetics at the surface. In addition, as concluded from the Tafel fitting analysis in Fig. 6c, the Tafel slope of the sample FeCo@NGHS (233.5 mV dec⁻¹) is relatively lower than those of the Fe@NGHS (323.96 mV dec⁻¹), Co@NGHS (278.7 mV dec⁻¹), Fe&Co@NGHS (265.21 mV dec⁻¹), FeCo@GHS (245.83 mV dec⁻¹), NGHS (323.68 mV dec⁻¹) and RuO₂ (271.06 mV dec⁻¹), which are consistent with the results of R_ct comparison. It reflects the excellent catalytic kinetics, and also indicates that the graphite structure and carbon matrix of the catalyst can enhance the electronic conductivity of the supported Fe/Co alloy.

Fig. 6 Assessment of the OER capabilities of FeCo@NGHS, Fe@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS, NGHS, and RuO₂ in 0.1 M KOH: (a) OER polarization curves, (b) Tafel slopes, (c) EIS Nyquist plots, and (d) stability test of FeCo@NGHS and RuO₂ maintained at 10 mA cm⁻².

The CV tests for the catalysts were performed at different scan rates (20–100 mV s⁻¹) in the faradaic region, and the C_dl data (shown in Fig. S5†) were obtained to evaluate the ECSA of the catalysts due to the fact that these two have a linear relationship.⁵⁶ As shown in Fig. S5g,† the calculated C_dl values of FeCo@NGHS, Fe@NGHS, Co@NGHS, Fe&Co@NGHS, FeCo@GHS and NGHS are 11.43 mF cm⁻², 11.19 mF cm⁻², 8.88 mF cm⁻², 10.2 mF cm⁻², 5.97 mF cm⁻² and 7.31 mF cm⁻², respectively. The FeCo@NGHS exhibits the largest C_dl value, which indicates that FeCo@NGHS has a larger ECSA and can create more abundant active sites for the OER process.¹²

The performance of FeCo@NGHS is significantly better than that of Fe&Co@NGHS. This is due to the regulation of the reaction rate between potassium ferricyanide and cobalt ions by the complexing agent sodium citrate. This enables the sample FeCo@NGHS to have appropriate alloy loading and form a large number of highly active Fe(Co)N_x active sites, leading to better OER performance. Due to the lack of sufficient doped N atoms, the lowest content of alloy loading, N atom doping and formed Fe(Co)N_x active sites resulted in the worst OER performance of FeCo@GHS in the alloy-loaded samples. Furthermore, the OER activity and kinetics of FeCo@NGHS are superior to those of the single-metal samples Fe@NGHS and Co@NGHS, which originate from the synergistic effect of Fe/Co alloys.

In the chronopotentiometry test (as shown in Fig. 6d), the potential of FeCo@NGHS hardly changed after working continuously for 63 h in 0.1 M KOH electrolyte with a current density of 10 mA cm⁻². Under the same conditions, the potential decay of RuO₂ is obvious (≈77%). The corresponding morphology and elemental composition of FeCo@NGHS were characterized after 63 h of OER testing. As shown by the SEM and TEM in Fig. S3d–f,† the FeCo@NGHS sample still has the morphology of uniform graphene spheres, and there are well-dispersed alloy NPs. According to the SAED (inset of Fig. S3e†), the (110), (200) and (211) planes belong to the Fe/Co alloy still clearly visible.

The high bifunctional performance of the sample FeCo@NGHS originates from the controllable loading of the Fe/Co alloy: (1) the complexation of sodium citrate constrains the combination of cobalt ions and potassium ferricyanide, resulting in the formation of more Fe(Co)N_x active sites; (2) melamine plays the main role in N doping and the following riveting of the alloy particles. It also has a certain role in forming hollow graphene spheres with high specific surface area. For FeCo@NGHS, the doped N atoms with high electronegativity will affect the adjacent carbon atoms, presenting a higher density of positive charges. It facilitates the chemical adsorption of oxygen molecules and diminishes the strength of the O–O bond, thereby accelerating the rate of the ORR.^57,58 In addition, the Fe(Co)N_x active sites, formed by the doped N atoms and the subsequent loading of the obtained alloy particles, play a key role in the improvement of the ORR and OER performance; (3) the material is based on graphene hollow spheres. The graphite substrate has high electronic conductivity and abundant structural defects. Graphene presents a three-dimensional spherical structure, which can accelerate the mass transfer of the reactants and reaction intermediates, and provide more active sites.

The bifunctionality of catalysts is typically assessed by quantifying the potential gap (ΔE) between the potential required for achieving 10 mA cm⁻² during the OER process and the half-wave potential observed during the ORR reaction.⁵⁹ The ΔE value of FeCo@NGHS is 0.775 V, which is lower than that of commercial Pt/C + RuO₂ (0.87 V) (Fig. 5h). It proves that the catalyst FeCo@NGHS has excellent bifunctional activity, and the bifunctional activity of the sample is comparable to metal-based carbonaceous catalysts that have been previously reported in many studies (see Table S4†).

Studies of the liquid electrolyte and all-solid-state flexible ZABs

A rechargeable liquid ZAB utilizing FeCo@NGHS was constructed, employing a polished zinc sheet as the anode material and an air cathode coated with the catalyst. It was tested for ZAB performance. As shown in Fig. 7a, the FeCo@NGHS-assembled ZAB has a slightly higher open circuit voltage (OCV) (1.51 V) than the commercial catalyst-assembled ZAB (1.50 V). The FeCo@NGHS-based ZAB achieved a maximum power density of 121.19 mW cm⁻², surpassing that of the Pt/C + RuO₂-based ZAB, which was 102.63 mW cm⁻² (as illustrated in Fig. 7b), and is higher than those of some advanced bifunctional electrocatalysts reported recently (Table S7†). Fig. 7c shows the specific capacity curve of the liquid ZAB. The specific capacitance of the liquid ZAB was calculated and analyzed by normalizing the mass loss of the zinc flakes. At constant operating currents of 50, 100, and 200 mA cm⁻², the specific capacities of FeCo@NGHS-based ZAB reaches 802, 800, and 786 mA h g_Zn⁻¹, which are close to the theoretical specific capacity (820 mA h g_Zn⁻¹). In Fig. 7d, when the FeCo@NGHS-based ZAB is charged and discharged to a current density of 10 mA cm⁻², the voltage reaches 1.95 V and 1.28 V, respectively. Furthermore, the charge–discharge potential difference (0.67 V) is lower than that of Pt/C + RuO₂ (0.72 V), showing a better reversibility.

Fig. 7 Evaluation of the liquid ZAB performance utilizing FeCo@NGHS and a 20 wt% Pt/C + RuO₂ composite: (a) OCV measurement. (b) Discharge polarization and power density profiles comparing FeCo@NGHS and 20 wt% Pt/C + RuO₂ (in a 1 : 1 mass ratio). (c) Analysis of the specific discharge capacities at 50, 100, and 200 mA cm⁻². (d) Charge–discharge polarization curves. (e) Galvanostatic cycling stability of the ZAB at 10 mA cm⁻² using FeCo@NGHS and the Pt/C + RuO₂ composite.

To investigate the cycling stability of the catalyst in the ZAB (as shown in Fig. 7e), the galvanostatic charge–discharge cycle test was performed at 10 mA cm⁻². The catalyst Pt/C + RuO₂ shows a voltage difference of 1.50 V after 210 cycles, while the energy efficiency decreased from 54.2% to 40.7%. The weakening of the stability may be due to the reduction of the interaction between Pt NPs and carbon materials, and the loss of RuO₂ activity.³⁹ Similarly, the electrocatalyst FeCo@NGHS-assembled ZAB only increases the voltage difference from the initial 1.01 V to 1.09 V after 210 cycles (∼33 h), and the energy efficiency only decreased from the initial 53.4% to 51.1%, which confirmed the excellent charge–discharge cycle stability. The superior performance could stem from the electrocatalyst's robust structural integrity and stable surface chemistry.

Based on the excellent performance of rechargeable liquid ZABs, a flexible ASS–ZAB was assembled based on FeCo@NGHS (inset of Fig. 8a). Fig. 8a shows the OCV (1.45 V) of the FeCo@NGHS-based ASS–ZABs, which is slightly higher than that of ASS–ZABs assembled with commercial Pt/C + RuO₂ (1.42 V). The discharge polarization curves and the corresponding power density curves are presented in Fig. 8b, in which the ASS–ZAB assembled with FeCo@NGHS catalyst exhibits a peak power density of 74.06 mW cm⁻². This is better than that of commercial Pt/C + RuO₂ (35.35 mW cm⁻²) and many advanced electrocatalysts in ASS–ZABs (shown in Table S7†). Fig. 8c displays the galvanostatic charge–discharge cycling curve at 1 mA cm⁻², with an initial discharge potential of 1.32 V and a charge potential of 1.96 V, respectively. The initial charge–discharge voltage difference of the battery is 0.64 V, which increases to 0.76 V by the 40^th cycle, corresponding to a decrease in energy efficiency from the initial 67.3% to 62.5%. To further study the flexibility of the ASS–ZAB, the battery was bent at different angles and simultaneously tested its charge–discharge cycle stability at 1 mA cm⁻², as shown in Fig. 8d. To be sure, the battery still maintains a certain discharge (1.28 V) and charge (1.95 V) plateau after being bent at any angle. In addition, the potentials gap and round-trip efficiency of the as-assembled battery show no significant change under various bending angle. Furthermore, the ASS–ZABs can exhibit a voltage retention of 98.3% when bended from initial 0° (1.448 V), 90° (1.431 V) to 180° (1.424 V), suggesting its good flexibility and stability.

Fig. 8 Utilizing FeCo@NGHS and the 20 wt% Pt/C + RuO₂ composite as air cathode catalysts in ASS–ZABs: (a) OCV assessment (inset: schematic of an ASS–ZAB), (b) discharge polarization and power density curves, (c) galvanostatic discharge–charge cycling at 1 mA cm⁻², (d) galvanostatic cycling stability of ASS–ZABs incorporating FeCo@NGHS under various bending conditions.

Conclusions

In this work, we report on the successfully developed aqueous-solvent one-step synthesis route for controllable loading of the Fe/Co alloy on heteroatom-doped hollow graphene spheres (FeCo@NGHS) realized by small molecule regulation. The complexation of sodium citrate and cobalt ions is used to buffer the formation of PBA and regulate the formation of Fe/Co alloy NPs. Through research, it was found that the introduction of sodium citrate resulted in the formation of more Fe(Co)N_x active sites in the catalyst material, resulting in a higher electrochemically active specific surface. As a nitrogen source, melamine can successfully dope N atoms into the hollow graphene sphere base material, which ensures the proper loading of transition metals, and provides the necessary conditions for the generation of Fe(Co)N_x active sites. Therefore, the introduction of sodium citrate and melamine is the key to ensure the controllable loading of alloy particles and form more Fe(Co)N_x active sites. The research results indicate that the designed catalyst material FeCo@NGHS demonstrates optimal bifunctional catalytic activity for ORR and OER in alkaline electrolytes. The FeCo@NGHS catalyst demonstrates a more favorable onset potential of 0.961 V and a half-wave potential of 0.846 V (vs. RHE) during the ORR, surpassing those of the commercial Pt/C catalyst. At a current density of 10 mA cm⁻², the OER overpotential required for the electrocatalyst is just 391 mV, outperforming the commercial RuO₂ catalyst in this regard. In terms of stability for both ORR and OER, this material also exhibits remarkable long-term stability in electrocatalytic performance and exceptional structural stability. Additionally, the rechargeable liquid ZAB and flexible ASS–ZAB assembled from FeCo@NGHS display high discharge power densities and enduring charge–discharge cycling performance. The FeCo@NGHS based air cathode exhibits outstanding performance in the flexible ZAB, presenting a high OCV (1.45 V) and peak power density (74.06 mW cm⁻²). This research presents a novel synthetic approach for developing outstanding bifunctional oxygen electrocatalysts, with implications for advancing clean energy storage and conversion technologies.

Data availability

The data provided in this investigation are available from the corresponding authors on reasonable request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge financial support from the Zhejiang Provincial Natural Science Foundation (no. LR22E070001), the National Natural Science Foundation of China (no. 12275239 and 11975205), the Guangdong Basic and Applied Basic Research Foundation (no. 2020B1515120048), the Fundamental Research Funds of Zhejiang Sci-Tech University (no. 23062096-Y), and the Fundamental Research Funds of Ningbo University of Technology (no. 2022TS17).

References

1. H. Yan, Y. Wang, Y. Xin, Z. Jiang, B. Deng and Z.-J. Jiang, Carbon nanotube support, carbon loricae and oxygen defect Co-promoted superior activities and excellent durability of RuO₂ nanoparticles towards the pH-universal H₂ evolution, Small, 2024, 20, 2406070.
2. D. Hu, Y. Wang, W. Chen, Z. Jiang, B. Deng and Z.-J. Jiang, Strong metal-support interaction modulation between Pt nanoclusters and Mn₃O₄ nanosheets through oxygen vacancy control to achieve high activities for acidic hydrogen evolution, Small, 2024, 20, 2402652.
3. H. Yan, Z. Jiang, B. Deng, Y. Wang and Z.-J. Jiang, Ultrathin carbon coating and defect engineering promote RuO₂ as an efficient catalyst for acidic oxygen evolution reaction with super-high durability, Adv. Energy Mater., 2023, 13, 2300152.
4. B. Jiang, J. Zhu, Z. Xia, J. Lyu, X. Li, L. Zheng, C. Chen, S. Chaemchuen, T. Bu, F. Verpoort, S. Mu, J. Wu, J. Wang and Z. Kou, Correlating single-atomic ruthenium interdistance with long-range interaction boosts hydrogen evolution reaction kinetics, Adv. Mater., 2024, 36, 2310699.
5. S. Chen, L. Ma, S. Wu, S. Wang, Z. Li, A. A. Emmanuel, M. R. Huqe, C. Zhi and J. A. Zapien, Uniform virus-like Co-N-Cs electrocatalyst derived from prussian blue analog for stretchable fiber-shaped Zn-air batteries, Adv. Funct. Mater., 2020, 30, 1908945.
6. Q. Lu, J. Yu, X. Zou, K. Liao, P. Tan, W. Zhou, M. Ni and Z. Shao, Self-catalyzed growth of Co, N-Codoped CNTs on carbon-encased CoSx surface: A noble-metal-free bifunctional oxygen electrocatalyst for flexible solid Zn-air batteries, Adv. Funct. Mater., 2019, 29, 1904481.
7. L. Ye, W. Chen, Z.-J. Jiang and Z. Jiang, Co/CoO heterojunction rich in oxygen vacancies introduced by O₂ plasma embedded in mesoporous walls of carbon nanoboxes covered with carbon nanotubes for rechargeable zinc–air battery, Carbon Energy, 2024, 6, e457.
8. D. Chen, J. Zhu, X. Mu, R. Cheng, W. Li, S. Liu, Z. Pu, C. Lin and S. Mu, Nitrogen-doped carbon coupled FeNi₃ intermetallic compound as advanced bifunctional electrocatalyst for OER, ORR and Zn-air batteries, Appl. Catal., B, 2020, 268, 118729.
9. D. Zha, R. Wang, S. Tian, Z.-J. Jiang, Z. Xu, C. Qin, X. Tian and Z. Jiang, Defect engineering and carbon supporting to achieve Ni-doped CoP₃ with high catalytic activities for overall water splitting, Nano-Micro Lett., 2024, 16, 250.
10. J. Zhang, W. Nie, R. Wang, C. Qin, Z.-J. Jiang, G.-r. Jin and Z. Jiang, Plasma technique regulates the electronic structure and dual functional catalytic performance of p-VNiCoP_y/NiFeO_x heterojunction catalysts for hydrogen production through overall water splitting, J. Alloys Compd., 2024, 1004, 175900.
11. J. Kuang, B. Deng, Z. Jiang, Y. Wang and Z.-J. Jiang, Sr-stabilized IrMnO₂ solid solution nano-electrocatalysts with superior activity and excellent durability for oxygen evolution reaction in acid media, Adv. Mater., 2024, 36, 2306934.
12. B. Chen, Z. Jiang, J. Huang, B. Deng, L. Zhou, Z.-J. Jiang and M. Liu, Cation exchange synthesis of Ni_xCo_(3-x)O₄ (x = 1.25) nanoparticles on aminated carbon nanotubes with high catalytic bifunctionality for the oxygen reduction/evolution reaction toward efficient Zn-air batteries, J. Mater. Chem. A, 2018, 6, 9517–9527.
13. H. Zhu, Y. Wang, Z. Jiang, B. Deng, Y. Xin and Z.-J. Jiang, Defect engineering promoted ultrafine Ir nanoparticle growth and Sr single-atom adsorption on TiO₂ nanowires to achieve high-performance overall water splitting in acidic media, Adv. Energy Mater., 2024, 14, 2303987.
14. Q. Ma and S. Mu, Acidic oxygen evolution reaction: Mechanism, catalyst classification, and enhancement strategies, Interdiscip. Mater., 2023, 2, 53–90.
15. Y. Liu, Z. Jiang and Z.-J. Jiang, Plasma-assisted formation of oxygen defective NiCoO/NiCoN heterostructure with improved ORR/OER activities for highly durable all-solid-state zinc-air batteries, Adv. Funct. Mater., 2023, 33, 2302883.
16. M. Yu, Z. Wang, C. Hou, Z. Wang, C. Liang, C. Zhao, Y. Tong, X. Lu and S. Yang, Nitrogen-doped Co₃O₄ mesoporous nanowire arrays as an additive-free air-cathode for flexible solid-state zinc-air batteries, Adv. Mater., 2017, 29, 1602868.
17. C. Tang, B. Wang, H. F. Wang and Q. Zhang, Defect engineering toward atomic Co-N_x-C in hierarchical graphene for rechargeable flexible solid Zn-air batteries, Adv. Mater., 2017, 29, 1703185.
18. Y.-N. Chen, Y. Guo, H. Cui, Z. Xie, X. Zhang, J. Wei and Z. Zhou, Bifunctional electrocatalysts of MOF-derived Co-N/C on bamboo-like MnO nanowires for high-performance liquid- and solid-state Zn-air batteries, J. Mater. Chem. A, 2018, 6, 9716–9722.
19. Y.-e. Feng, W. Chen, L. Zhao, Z.-J. Jiang, X. Tian and Z. Jiang, Ar/NH₃ plasma etching of cobalt-nickel selenide microspheres rich in selenium vacancies wrapped with nitrogen doped carbon nanotubes as highly efficient air cathode catalysts for zinc-air batteries, Small Methods, 2024, 2400565.
20. J. Chen, G. Fu, Y. Tian, X. Li, M. Luo, X. Wei, T. Zhang, T. Gao, C. Chen, S. Chaemchuen, X. Xu, X. Sun, T. Bu, F. Verpoort, J. Wang and Z. Kou, Three-dimensional-printed Ni-based scaffold design accelerates bubble escape for ampere-level alkaline hydrogen evolution reaction, Interdiscip. Mater., 2024, 3, 595–606.
21. R. Nandan, H. R. Devi, R. Kumar, A. K. Singh, C. Srivastava and K. K. Nanda, Inner sphere electron transfer promotion on homogeneously dispersed Fe-N_x centers for energy-efficient oxygen reduction reaction, ACS Appl. Mater. Interfaces, 2020, 12, 36026–36039.
22. W. Zhu, Y. Pei, Y. Liu, J. Zhang, Y. Qin, Y. Yin and M. D. Guiver, Mass transfer in a Co/N/C catalyst layer for the anion exchange membrane fuel cell, ACS Appl. Mater. Interfaces, 2020, 12, 32842–32850.
23. F.-L. Meng, Z.-L. Wang, H.-X. Zhong, J. Wang, J.-M. Yan and X.-B. Zhang, Reactive multifunctional template-induced preparation of Fe-N-doped mesoporous carbon microspheres towards highly efficient electrocatalysts for oxygen reduction, Adv. Mater., 2016, 28, 7948–7955.
24. F. Meng, H. Zhong, D. Bao, J. Yan and X. Zhang, In situ coupling of strung Co4N and intertwined N-C fibers toward free-standing bifunctional cathode for robust, efficient, and flexible Zn-air batteries, J. Am. Chem. Soc., 2016, 138, 10226–10231.
25. S. Du, Z. Ren, Y. Qu, J. Wu, W. Xi, J. Zhu and H. Fu, Co₃O₄ nanosheets as a high-performance catalyst for oxygen evolution proceeding via a double two-electron process, Chem. Commun., 2016, 52, 6705–6708.
26. X. Liu, I. S. Amiinu, S. Liu, K. Cheng and S. Mu, Transition metal/nitrogen dual-doped mesoporous graphene-like carbon nanosheets for the oxygen reduction and evolution reactions, Nanoscale, 2016, 8, 13311–13320.
27. F. M. Sapountzi, J. M. Gracia, C. J. Weststrate, H. O. A. Fredriksson and J. W. Niemantsverdriet, Electrocatalysts for the generation of hydrogen, oxygen and synthesis gas, Prog. Energy Combust. Sci., 2017, 58, 1–35.
28. C. Tang, H. F. Wang and Q. Zhang, Multiscale principles to boost reactivity in gas-involving energy electrocatalysis, Acc. Chem. Res., 2018, 51, 881–889.
29. J. Sun, S. E. Lowe, L. Zhang, Y. Wang, K. Pang, Y. Wang, Y. Zhong, P. Liu, K. Zhao, Z. Tang and H. Zhao, Ultrathin nitrogen-doped holey carbon@graphene bifunctional electrocatalyst for oxygen reduction and evolution reactions in alkaline and acidic media, Angew. Chem., Int. Ed., 2018, 57, 16511–16515.
30. M. Jahan, Q. Bao and K. P. Loh, Electrocatalytically active graphene-porphyrin MOF composite for oxygen reduction reaction, J. Am. Chem. Soc., 2012, 134, 6707–6713.
31. Q. Zuo, G. Cheng and W. Luo, A reduced graphene oxide/covalent cobalt porphyrin framework for efficient oxygen reduction reaction, Dalton Trans., 2017, 46, 9344–9348.
32. L. Wei, L. Qiu, Y. Liu, J. Zhang, D. Yuan and L. Wang, Mn-doped Co-N-C dodecahedron as a bifunctional electrocatalyst for highly efficient Zn-air batteries, ACS Sustainable Chem. Eng., 2019, 7, 14180–14188.
33. C. Deng, K. H. Wu, J. Scott, S. Zhu, X. Zheng, R. Amal and D. W. Wang, Spherical murray-type assembly of Co-N-C Nanoparticles as a high-performance trifunctional electrocatalyst, ACS Appl. Mater. Interfaces, 2019, 11, 9925–9933.
34. Z. Liang, H. Zheng and R. Cao, Recent advances in Co-based electrocatalysts for the oxygen reduction reaction, Sustainable Energy Fuels, 2020, 4, 3848–3870.
35. L. Wang, J. Zhang, L. Zheng, J. Yang, Y. Li, X. Wan, X. Liu, X. Zhang, R. Yu and J. Shui, Carbon black-supported FM-N-C (FM = Fe, Co, and Ni) single-atom catalysts synthesized by the self-catalysis of oxygen-coordinated ferrous metal atoms, J. Mater. Chem. A, 2020, 8, 13166–13172.
36. T. Palaniselvam, V. Kashyap, S. N. Bhange, J.-B. Baek and S. Kurungot, Nanoporous graphene enriched with Fe/Co-N active sites as a promising oxygen reduction electrocatalyst for anion exchange membrane fuel cells, Adv. Funct. Mater., 2016, 26, 2150–2162.
37. C. Lai, J. Fang, X. Liu, M. Gong, T. Zhao, T. Shen, K. Wang, K. Jiang and D. Wang, In situ coupling of NiFe nanoparticles with N-doped carbon nanofibers for Zn-air batteries driven water splitting, Appl. Catal., B, 2021, 285, 119856.
38. H.-L. Jia, J. Zhao, L. Gu, Z.-J. Peng, Z.-L. Bao, X.-L. Sun and M.-Y. Guan, Highly active Co-N-doped graphene as an efficient bifunctional electrocatalyst (ORR/HER) for flexible all-solid-state zinc-air batteries, Sustainable Energy Fuels, 2020, 4, 6165–6173.
39. X. Liu, L. Wang, P. Yu, C. Tian, F. Sun, J. Ma, W. Li and H. Fu, A stable bifunctional catalyst for rechargeable zinc-air batteries: iron-cobalt nanoparticles embedded in a nitrogen-doped 3D carbon matrix, Angew. Chem., Int. Ed., 2018, 57, 16166–16170.
40. X. Hao, Z. Jiang, B. Zhang, X. Tian, C. Song, L. Wang, T. Maiyalagan, X. Hao and Z. J. Jiang, N-doped carbon nanotubes derived from graphene oxide with embedment of FeCo nanoparticles as bifunctional air electrode for rechargeable liquid and flexible all-solid-state zinc-air batteries, Adv. Sci., 2021, 8, 2004572.
41. L. Chen, H. Jiang, H. Jiang, H. Zhang, S. Guo, Y. Hu and C. Li, Mo-based ultrasmall nanoparticles on hierarchical carbon nanosheets for superior lithium ion storage and hydrogen generation catalysis, Adv. Energy Mater., 2017, 7, 1602782.
42. Z. Guo, F. Wang, Y. Xia, J. Li, A. G. Tamirat, Y. Liu, L. Wang, Y. Wang and Y. Xia, In situ encapsulation of core-shell-structured Co@Co₃O₄ into nitrogen-doped carbon polyhedra as a bifunctional catalyst for rechargeable Zn-air batteries, J. Mater. Chem. A, 2018, 6, 1443–1453.
43. K. Michal and J. Mietek, Gas adsorption characterization of ordered organic-inorganic nanocomposite materials, Chem. Mater., 2001, 13, 3169–3183.
44. H. Jiang, J. Gu, X. Zheng, M. Liu, X. Qiu, L. Wang, W. Li, Z. Chen, X. Ji and J. Li, Defect-rich and ultrathin N doped carbon nanosheets as advanced trifunctional metal-free electrocatalysts for the ORR, OER and HER, Energy Environ. Sci., 2019, 12, 322–333.
45. L. Liu, F. Yan, K. Li, C. Zhu, Y. Xie, X. Zhang and Y. Chen, Ultrasmall FeNi₃N particles with an exposed active (110) surface anchored on nitrogen-doped graphene for multifunctional electrocatalysts, J. Mater. Chem. A, 2019, 7, 1083–1091.
46. H.-J. Niu, S.-S. Chen, J.-J. Feng, L. Zhang and A.-J. Wang, Assembled hollow spheres with CoFe alloyed nanocrystals encapsulated in N, P-doped carbon nanovesicles: An ultra-stable bifunctional oxygen catalyst for rechargeable Zn-air battery, J. Power Sources, 2020, 475, 228594.
47. J. Zhang, M. Zhang, Y. Zeng, J. Chen, L. Qiu, H. Zhou, C. Sun, Y. Yu, C. Zhu and Z. Zhu, Single Fe atom on hierarchically porous S, N-codoped nanocarbon derived from porphyra enable boosted oxygen catalysis for rechargeable Zn-air batteries, Small, 2019, 15, 1900307.
48. M. Li, T. Liu, X. Bo, M. Zhou and L. Guo, A novel flower-like architecture of FeCo@NC-functionalized ultra-thin carbon nanosheets as a highly efficient 3D bifunctional electrocatalyst for full water splitting, J. Mater. Chem. A, 2017, 5, 5413–5425.
49. H. Lei, Z. Wang, F. Yang, X. Huang, J. Liu, Y. Liang, J. Xie, M. S. Javed, X. Lu, S. Tan and W. Mai, NiFe nanoparticles embedded N-doped carbon nanotubes as high-efficient electrocatalysts for wearable solid-state Zn-air batteries, Nano Energy, 2020, 68, 104293.
50. Z. Wang, J. Ang, B. Zhang, Y. Zhang, X. Y. D. Ma, T. Yan, J. Liu, B. Che, Y. Huang and X. Lu, FeCo/FeCoNi/N-doped carbon nanotubes grafted polyhedron-derived hybrid fibers as bifunctional oxygen electrocatalysts for durable rechargeable zinc-air battery, Appl. Catal., B, 2019, 254, 26–36.
51. J. Zhang, M. Zhang, L. Qiu, Y. Zeng, J. Chen, C. Zhu, Y. Yu and Z. Zhu, Three-dimensional interconnected core-shell networks with Ni(Fe)OOH and M-N-C active species together as high-efficiency oxygen catalysts for rechargeable Zn-air batteries, J. Mater. Chem. A, 2019, 7, 19045–19059.
52. X. Hao, Z. Jiang, X. Tian, X. Hao, T. Maiyalagan and Z.-J. Jiang, A novel particle-in-nanoplate architecture of iron nickel phosphide intertwined with carbon nanotubes for efficient water oxidation and high-performance sodium-ion batteries, J. Alloys Compd., 2019, 791, 1220–1230.
53. K. P. Singh, E. J. Bae and J. S. Yu, Fe-P: a new class of electroactive catalyst for oxygen reduction reaction, J. Am. Chem. Soc., 2015, 137, 3165–3168.
54. Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier and H. Dai, Co₃O₄ nanocrystals on graphene as a synergistic catalyst for oxygen reduction reaction, Nat. Mater., 2011, 10, 780–786.
55. C. H. Choi, H. K. Lim, M. W. Chung, J. C. Park, H. Shin, H. Kim and S. I. Woo, Long-range electron transfer over graphene-based catalyst for high-performing oxygen reduction reactions: importance of size, N-doping, and metallic impurities, J. Am. Chem. Soc., 2014, 136, 9070–9077.
56. T. Liu, S. Cai, Z. Gao, S. Liu, H. Li, L. Chen, M. Li and H. Guo, Facile synthesis of the porous FeCo@nitrogen-doped carbon nanosheets as bifunctional oxygen electrocatalysts, Electrochim. Acta, 2020, 335, 135647.
57. T. Xing, Y. Zheng, L. H. Li, B. C. C. Cowie, D. Gunzelmann, S. Z. Qiao, S. Huang and Y. Chen, Observation of active sites for oxygen reduction reaction on nitrogen-doped multilayer graphene, ACS Nano, 2014, 8, 6856–6862.
58. D. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo and J. Nakamura, Active sites of nitrogen-doped carbon materials for oxygen reduction reaction clarified using model catalysts, Science, 2016, 351, 361–365.
59. F. L. Li, Q. Shao, X. Huang and J. P. Lang, Nanoscale trimetallic metal-organic frameworks enable efficient oxygen evolution electrocatalysis, Angew. Chem., Int. Ed., 2018, 57, 1888–1892.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4qi02166e

‡ These authors contributed equally.

---