MnO–nitrogen doped graphene as a durable non-precious hybrid catalyst for the oxygen reduction reaction in anion exchange membrane fuel cells

Abstract

A promising non-precious metal catalyst containing manganese oxide and N-graphene is synthesized and is recognised as an efficient and durable electrocatalyst for the oxygen reduction reaction (ORR) in an alkaline medium. Hydrothermally synthesized Mn₃O₄ is incorporated into graphene mixed with melamine as a nitrogen source and annealed at a temperature between 700 °C and 1000 °C. The phase formation, purity and morphological behaviours of pre-formed Mn₃O₄ and MnO after annealing are systematically studied by X-ray diffraction (XRD), high resolution scanning electron microscopy (HR-SEM) and transmission electron microscopy (TEM). The MnO and N-doped graphene (MnO/NG) catalyst enhances ORR activity compared to other combinations of hybrid catalysts, such as Mn₃O₄/G and MnO/G, as observed from the linear sweep voltammograms. It is found that MnO/NG annealed at 900 °C demonstrates excellent ORR catalytic activity in O₂ saturated 0.1 M aqueous KOH electrolyte with a dominant 4e⁻ transfer process, as confirmed by both RDE and RRDE measurements. The stability and durability of this optimized catalyst is ascertained by potential cycling between −0.8 and 0.2 V vs. Ag/AgCl up to 10 000 potential cycles; superior durability was found in comparison to the state-of-art Pt/C catalyst. This MnO/NG-900 hybrid catalyst is used as a cathode catalyst, and we fabricate a membrane electrode assembly (MEA) to validate the catalyst in an anion exchange membrane fuel cell (AEMFC). A peak power density of 13 mW cm⁻² at 30 °C under ambient pressure is realized; thus, our catalyst appears to be promising as an alternative non-precious metal catalyst for AEMFCs.

---

Introduction

Due to rapid technological developments, energy demands have increased extensively, which has stimulated intense research into alternative energy conversion and storage devices. In the past few decades, numerous advanced materials with high energies and power densities have been developed for energy devices such as fuel cells and batteries.^1,2 Oxygen reduction is a critical and sluggish electrochemical reaction in polymer electrolyte membrane fuel cells (PEMFCs) in both alkaline and acid media.³ Generally, Pt-based electrocatalysts are used in PEMFC electrodes for efficient oxygen reduction reaction (ORR) and to enhance the power density.^4,5 In order to enhance ORR activity and to improve Pt utilisation, optimization of Pt particle size and distribution on the support materials is required. In order to achieve this, several advanced surface functionalization processes to deposit Pt catalysts are being developed using bio-molecules,⁶ macromolecules,^7,8 conducting polymers,^9,10 etc. However, high cost, scarcity and long-run stability, especially at elevated temperature and low pH, are limitations of Pt-based electrocatalysts which limit their large-scale market opportunities.⁵ The demand for anion exchange membrane fuel cells (AEMFCs) has increased due to the fact that higher efficiency can be obtained without the use of precious metal catalysts because the over-potential of oxygen reduction is significantly lower in alkaline environments compared to the ORR in acidic media.¹¹

In recent years, graphene doped with heteroatom(s) alone or in combination with transition metals has attracted much attention for its remarkable electrochemical behaviour.^12–18 The concept of heteroatom doping on carbon and its interaction with transition metals to enhance ORR activity was first explored by Gupta et al.¹⁹ The heteroatoms, such as nitrogen, act as binding sites for transition metal ions and therefore provide active catalytic centres for ORR. Although N-doped graphene shows appreciable enhancement of ORR activity in both alkaline and acidic media, it still has yet to achieve fuel cell performance on par with the state-of-the-art Pt/C electrocatalyst.

In recent years, metal oxides such as Co₃O₄,²⁰ CoO,²¹ Fe₃O₄,²² MnO₂,^23,24 Mn₂O₃,²⁵ and Mn₃O₄ ^26,27 supported on carbon have been studied extensively as potential ORR electrocatalysts in alkaline medium. Among these metal oxides, manganese oxides have received greater attention due to their low cost, availability, and chemical and electrochemical stability.^28–32 In addition, manganese oxide exhibits excellent ORR activity towards the disproportionation reaction in which HO₂⁻ is converted to O₂ and OH⁻.^33,34 Recently, Zhang et al. demonstrated flower-like structured MnO₂ on reduced graphene oxide (RGO) and studied its synergistic effect towards the enhancement of electrocatalytic ORR activity in alkaline medium.²³ Very recently, Wu et al. revealed the ORR activity of a hybrid catalyst containing Mn₃O₄–graphene oxide nanoribbons (GONR), wherein the origin of the improved activity was explained and synergism between the metal oxide and GONR was identified.²⁶ In another report, Chen et al. developed a three-dimensional nitrogen-doped reduced graphene oxide/manganese oxide composite and demonstrated it to be a high-performing ORR electrocatalyst.³⁵ The improvements in the ORR activity and stability were attributed to the stable three-dimensional composite structure of the catalyst material. On the other hand, the stability of the Pt/C based state-of-the-art electrocatalyst is another challenging issue that must be addressed. A recent report by A. Zadick et al. indicates that Pt/C catalyst is prone to enormous instability in alkaline medium compared to acidic medium, with a loss of ∼60% of ECSA for just a short test of 150 potential cycles.³⁶ Hence, there is a need to develop durable electrocatalysts as alternatives to the traditional Pt/C catalyst in alkaline media.

Herein, we report a facile synthesis of manganese oxide nanostructures and their incorporation into nitrogen-doped graphene (MnO/NG) hybrid electrocatalysts towards ORR in alkaline media. The MnO/NG hybrids were synthesized by a two-step method; in the first step, Mn₃O₄ was synthesized at 160 °C by a hydrothermal method, while in the second step, the metal oxide was suspended in graphene in the presence or absence of nitrogen precursor, followed by an annealing process at temperatures between 800 °C and 1000 °C. The hybrid catalysts were assessed using a rotating disk electrode (RDE) and a rotating rink-disk electrode (RRDE) for systematic investigation of their electrochemical properties, with special emphasis on ORR catalytic activity and stability. The optimised MnO/NG-900 catalyst was subjected to 10 000 repeated potential cycles in alkaline medium; a ∼44 mV drop in its half-wave potential and a ∼6% decrease of the limiting current density were found. However, commercial Pt/C catalyst shows a remarkable decrease of its half-wave potential after 5000 repeated potential cycles. The optimized MnO/NG-900 hybrid catalyst was also used as a cathode catalyst in the fabrication of a membrane electrode assembly (MEA), and its performance was evaluated in a real fuel cell configuration. A peak power density of ∼13 mW cm⁻² was achieved in an AEMFC at ambient temperature and pressure. We are hopeful that the present findings will provide new insights and understanding for the synthesis of more advanced non-precious metal catalysts for efficient reduction of O₂, which is useful in AEMFCs.

Experimental section

Materials

Potassium permanganate (KMnO₄) and ethanol were purchased from Merck Chemicals, India and were used as received. Graphene nanopowder was procured from Angstron Materials Inc., USA. Pt/C (40 wt% Pt on carbon) was acquired from Alfa Aesar (Johnson Matthey Ltd.). The ammonium-type anion exchange membrane (AHA-NEOSEPTA), wherein tetra-alkyl ammonium is the cation group bonded to a polyolefin backbone chain, was purchased from ASTOM Corporation, Japan. Nafion (5 wt%) ionomer was purchased from Du Pont, USA, and Fumion-FAA-3 (10 wt%) ionomer solution (polyaromatic backbone with terminal quaternary ammonium ions and bromide counter-ions) was procured from Fumatech, Germany. All the solutions were prepared using de-ionized water (DI) (18.2 MΩ cm, Milli Q system).

Synthesis of manganese oxide

Manganese oxide was synthesized via a simple hydrothermal route, and this is schematically represented in step 1 of Scheme 1. In a typical run, 2 g of KMnO₄ was dissolved in 60 mL of 1 : 1 water/ethanol mixture followed by ultra-sonication for 1 h. The pink solution became a brown suspension upon sonication. Then, the above suspension was transferred to a 100 mL Teflon-lined stainless steel autoclave for hydrothermal reaction at 160 °C for 12 h. The obtained brown precipitate was filtered, washed thoroughly with DI water and dried at 80 °C overnight.

Scheme 1 Step 1 represents the synthesis process of manganese oxide at 160 °C by a hydrothermal process. Step 2 shows the annealing process of manganese oxide and graphene/N-graphene at different temperatures (800 °C, 900 °C and 1000 °C).

Synthesis of manganese oxide/nitrogen doped graphene hybrid catalysts

Manganese oxide/N-doped graphene (MnO/NG) hybrid catalysts were synthesized by heat treating melamine, graphene and manganese oxide in inert atmosphere under different temperatures (800 °C, 900 °C and 1000 °C), as shown in step 2 of Scheme 1. In a typical synthesis, the required amount of melamine was added to 20 mL of graphene slurry (5 mg/1 mL, 1 : 1 water/ethanol mixture) and ultra-sonicated for 1 h. Then, the appropriate amount of as-synthesized manganese oxide was added to the admixture and ultra-sonicated for another 1 h. The resulting mixture was dried overnight at 80 °C, and the residue was placed in a tubular furnace to conduct the annealing process at different temperatures (viz., 800 °C, 900 °C and 1000 °C) for 1 h under nitrogen atmosphere to obtain the MnO/NG hybrid catalysts. In order to obtain an optimized ORR electro-catalyst, the nitrogen content in the hybrid catalyst was varied (5, 10 and 15 wt%) and electrochemically investigated. A control experiment was also carried out by annealing the metal oxide/graphene mixture without a N-source at 900 °C.

Physicochemical characterization

Powder X-ray diffraction (XRD) was used to investigate the crystal structure of manganese oxide and was performed with a BRUKER D8 Advance diffractometer using Cu-Kα as the X-ray source (1.54 Å). High resolution scanning electron microscope (HR-SEM) images were recorded with a Hitachi instrument to visualize the morphologies of the manganese oxide and hybrid catalysts. An energy dispersive X-ray microanalyzer (OXFORD ISI 300 EDAX) attached to the electron microscope was used for elemental mapping of the hybrid materials. To identify the particle size distributions on graphene layers, transmission electron microscope (TEM) images were acquired using a Tecnai-20 G2 instrument operated at 200 kV. To identify elements and their electronic features, X-ray photoelectron spectra (XPS) were recorded with a MULTILAB 2000 XPS system, Thermo Scientific, using a twin anode Mg/Al (300/400 W) X-ray source in a 15 keV/150 W system.

Electrochemical characterization

The electrochemical tests were carried out using a potentiostat/galvanostat (Biologic, VSP/VMP 3B-20) at 25 °C in a standard three electrode system. A glassy carbon tip (diameter: 5 mm) was used as a substrate to electrochemically characterize the electrocatalysts developed in the present study. Prior to sample coating, the GC electrode was polished using 0.3 μm alumina powder and then washed ultrasonically in water and ethanol subsequently to obtain a clean and smooth electrode surface. 7.84 mg of each catalyst were added to 20 μL of Nafion ionomer and were dispersed in 1 mL of ethanol : water mixture (1 : 4) by ultra-sonication for 30 min. From the above dispersion, 15 μL was dropped onto the GC electrode surface and dried at room temperature. A catalyst loading of ∼0.6 mg cm⁻² was maintained on the GC electrode surface for all the samples. Pt wire and saturated calomel electrode (SCE) were used as the counter and reference electrodes, respectively. All potentials in this work are referenced with respect to the reversible hydrogen electrode (RHE). The RHE calibration is shown in the ESI.†

The electrochemical characterizations were carried out using cyclic voltammograms (CVs) and linear sweep voltammograms (LSVs) using 0.1 M aqueous KOH electrolyte. The CVs were recorded in N₂/O₂ saturated electrolyte between −0.8 and 0.2 V at a scan rate of 50 mV s⁻¹. To study the ORR kinetics, LSVs were performed between −0.8 and 0.1 V on a rotating working electrode operating under different rotational speeds (viz., 400, 800, 1200, 1600 and 2000 rpm) in O₂-saturated electrolyte at a scan rate of 5 mV s⁻¹. To evaluate the electrochemical stability of the catalysts, the CVs were repeated for 10 000 potential cycles in O₂ saturated electrolyte at a scan rate of 50 mV s⁻¹. After 10 000 continuous cycles, the LSVs were recorded at a scan rate of 5 mV s⁻¹ with a rotational rate of 1600 rpm. The durability results were compared with the state-of-the-art Pt/C (40 wt%). Rotating ring-disk electrode (RRDE) experiments with GC disks 5 mm in diameter and a Pt ring (Pine Instruments) were performed to validate the percentage of peroxide yield and the number of electrons transferred during the ORR process. To prepare the RRDE working electrode, 15 μL of as-prepared catalyst slurry was dropped onto the GC disk and allowed to dry at room temperature. SCE and Pt wire were used as the reference and counter electrodes, respectively.

Fabrication of membrane electrode assembly (MEA) and AEMFC performance evaluation

Commercial gas diffusion layers (GDL, SGL DC – 35) were used as backing layers for both the anode and cathode electrodes. To prepare the cathode electrodes, the optimised MnO/NG-900 catalysts were dispersed separately in ethanol followed by the addition of 10 wt% Fumion FAA-3 ionomer with continuous sonication for 30 min. The resulting slurries were brush-coated on to GDL until 2 mg cm⁻² catalyst loading was achieved. To prepare the anodes, commercial Pt/C (40 wt%) was used, and the above procedure was followed until a Pt loading of 0.5 mg cm⁻² was achieved. Both electrodes were immersed in 1 M aqueous KOH solution for 12 h for the exchange of Br⁻ ions from the ionomer with OH⁻. Commercial Fumatech membranes with a thickness of 50 μm were used as anion exchange membranes for making the MEAs. The membrane was treated with 2 M aqueous KOH at room temperature for 48 h in order to replace Cl⁻ ions in the membrane with OH⁻ ions.³⁷ The MEAs were obtained by sandwiching Fumatech membranes between the electrodes and hot pressing at 90 °C at a pressure of 20 kg cm⁻². For comparison, an MEA was prepared with commercial Pt/C (40 wt%) on both the anode and cathode. The MEAs were coupled with Teflon gas sealing gaskets and placed in single fuel cell text fixtures (Fuel Cell Tech. Inc., USA) with 5 cm² active areas and a parallel serpentine flow field machined on graphite plates. Gaseous H₂ and O₂ were fed into the anode and cathode sides of the cell, respectively, at a flow rate of 200 mL min⁻¹ through bubble humidifiers to maintain maximum relative humidity in the cell (ca. 100% RH). Galvanostatic polarization studies were performed using LCN100-36 electronic loads from Bitrode Corporation, USA. All experiments were carried out at 30 °C under ambient pressure.

Results and discussion

Synthesis of manganese oxide

Manganese oxide nanostructures were synthesised by a simple hydrothermal process which involves the formation of a hydrated metal oxide intermediate during sonication and the subsequent formation of manganese oxide nanostructures. After a subsequent hydrothermal process at 160 °C, the hydrated metal oxides orient themselves and form polyhedral Mn₃O₄ phase. The crystallinity of the as-synthesized Mn₃O₄ is examined by powder X-ray diffraction (XRD), and the results are shown in Fig. 1a. A well-defined characteristic profile of the hausmannite crystal structure is observed; the corresponding diffraction peaks can be indexed to PDF #04-007-1814, which corresponds to pure Mn₃O₄ phase. The lattice constants were calculated from the diffraction patterns and were found to be a = 5.7650 Å and c = 9.4395 Å; these results are also in good agreement with the above PDF reference. The polyhedral structures of the Mn₃O₄ nano-crystals can also be clearly observed in the HR-SEM and TEM images, as shown in Fig. 1b and c.

Fig. 1 (a) XRD pattern of hydrothermally synthesized manganese oxide, confirming the Mn₃O₄ phase, (b) HR-SEM image showing the morphology of the Mn₃O₄ particles and (c) TEM image of the Mn₃O₄ particles.

To synthesize the manganese oxide/N-graphene hybrid catalysts, a mixture of Mn₃O₄, graphene and melamine is heat treated at a temperature between 700 °C and 1000 °C to facilitate melamine decomposition and N doping onto the graphene matrix. Furthermore, the presence of electronegative nitrogen provides favourable anchoring sites for the metal oxides to co-ordinate with the graphene layers and enables strong coupling between Mn and graphene; this can confer synergistic effects on the electrochemical properties of the hybrid catalyst, as explained in the following sections.²⁰ During the annealing process, it is noted that the initial pure Mn₃O₄ phase is retained in all the samples that are heat treated up to 700 °C (Fig. 2a). However, the sample heat treated at 800 °C shows the diffraction peaks associated with Mn₃O₄ and MnO, demonstrating the presence of a mixed phase with predominant MnO phase. Nevertheless, samples annealed at 900 °C and 1000 °C, only the diffraction peaks associated with MnO phase are observed. From the above observations, it is noted that the temperature of 900 °C appears to be the ideal transition temperature for the phase transformation of Mn₃O₄ to MnO. The samples annealed at 700, 800, 900 and 1000 °C are hereafter denoted as Mn₃O₄/NG-700, MnO/NG-800, MnO/NG-900 and MnO/NG-1000, respectively. The X-ray diffraction peaks of MnO/NG-800, MnO/NG-900 and MnO/NG-1000 (see Fig. 2a) are positioned at 35.0, 40.5, 58.7, 70.0 and 73.9°, corresponding to the (1 1 1), (2 0 0), (2 2 0), (3 1 1) and (2 2 2) planes, respectively. They are cubic in nature (PDF #01-075-6876), with the Fm3̄m space group (225). In the case of MnO/NG-800, the peaks observed at 31.1 and 50.2° are attributed to the (2 0 0) and (2 0 4) planes of Mn₃O₄, respectively. However, for the MnO/NG-900 and MnO/NG-1000 hybrids, these peaks are not observed, indicating the complete transformation of Mn₃O₄ to MnO phase in the graphene layers. The resulting phase transformation is also seen in earlier reports, indicating that the manganese oxide phases are temperature-dependent and that the final reduced MnO phase is formed at higher temperatures.^38–40 As the annealing temperature increases, manganese oxide undergoes several reduction steps: MnO₂ → Mn₂O₃ → Mn₃O₄ → MnO. The crystallite sizes of MnO for the MnO/NG-800, MnO/NG-900 and MnO/NG-1000 hybrid catalysts are calculated by considering the high-intensity (2 0 0) peak using the Scherrer equation;⁴¹ they were found to be 36.6, 34.8 and 41.3 nm, respectively. Fig. S1a† shows the HR-SEM image for the MnO/NG-900 hybrid catalyst, wherein cubic MnO nanoparticles are embedded on the graphene matrix. The elemental mapping and distributions of C, N, Mn and O for MnO/NG-900 on the entire area of the image indicates the uniform distribution of MnO particles on N-graphene (see Fig. S1b–f†). TEM images of MnO/NG-800, MnO/NG-900 and MnO/NG-1000 are shown in Fig. 2b–d, and the average particle sizes are calculated to be 105, 90 and 130 nm, respectively. For MnO/NG-900, uniform distributions of MnO particles are observed on graphene sheets and the particles have quasi-spherical shapes. From the above XRD and microscopy images, 800 °C is perceived as a critical temperature for the phase transformation from Mn₃O₄ to MnO and hence for the variation in their particle sizes. However, at 1000 °C, particle growth is observed due to the higher annealing temperature of the hybrid material, and the particles are randomly distributed on the graphene matrix.

Fig. 2 (a) XRD patterns of manganese oxide/NG hybrid catalysts annealed at different temperatures (700 °C, 800 °C, 900 °C and 1000 °C). * represents Mn₃O₄ phase, JCPDS #04-007-1841. (b), (c) and (d) TEM images of the MnO/NG catalysts annealed at 800 °C, 900 °C and 1000 °C, respectively.

Electrochemical characterisations

The ORR electrocatalytic activities of all the electrocatalysts and their variations with respect to temperature are evaluated using CVs and LSVs, as shown in Fig. 3a and b. Well-defined reduction peaks are observed for the catalysts in the region between 0.60 and 0.65 V; these are attributed to the electrocatalytic reduction of oxygen. On the other hand, CVs recorded in N₂ saturated 0.1 M aqueous KOH electrolyte (Fig. 3a inset) show several redox peaks corresponding to the conversion process of Mn_xO_y → Mn(OH)₂ → MnOOH → Mn_xO_y. The aforementioned redox peaks are attributed to the disproportionation activity of manganese oxides.⁴² However, more intense reduction peaks are observed in O₂-saturated electrolyte; these are attributed to the electrocatalytic oxygen reduction reaction of the hybrid electrocatalysts. Fig. 3b shows LSVs of MnO/NG-800, MnO/NG-900 and MnO/NG-1000. The onset potential is one of the important parameters to evaluate ORR activity of electrocatalysts and it is determined from the intersection of the tangents between the baseline and the current. It can be seen that the onset potentials for MnO/NG-800, MnO/NG-900 and MnO/NG-1000 hybrid catalysts are 0.86, 0.89 and 0.85 V, respectively. The higher onset and half-wave potentials observed for the MnO/NG-900 catalyst indicate superior catalytic behaviour compared to the other electrocatalysts. The existence of Mn₃O₄ phase (see Fig. 2b) in MnO/NG-800 catalyst may be the cause of the decrease in the ORR activity. On the other hand, the MnO/NG-1000 catalyst, which was annealed at 1000 °C, shows a great reduction in its catalytic activity; this may be due to the larger crystallite size of the metal oxide particles. The phase purity and the optimal crystallite size for the MnO/NG-900 catalyst enhanced its ORR activity; hence, this catalyst was considered for further investigation in the following section. As the nitrogen content on graphene layers plays a critical role in ORR, therefore, the nitrogen percentage in MnO/NG-900 catalyst is optimized and the corresponding CVs and LSVs are shown in Fig. 3c and d, respectively. The onset potentials for MnO/NG-900 (5%), MnO/NG-900 (10%) and MnO/NG-900 (15%) hybrid catalysts are around 0.86, 0.89 and 0.89 V, respectively (Fig. 3d). Although the onset potentials of MnO/NG-900 (10%) and MnO/NG-900 (15%) are similar, a 10 mV negative shift in the half-wave potential region is observed for the latter. The comparative electrochemical parameters for all the catalysts studied above are provided in Table 1. In view of the superior electrochemical behaviour of the MnO/NG-900 (10%) hybrid catalyst in terms of the onset and half-wave potentials, its physicochemical and electrochemical behaviours are also investigated, and are discussed in the following sections.

Fig. 3 (a) CV and (b) LSVs of MnO/NG (annealed at 800 °C, 900 °C and 1000 °C) recorded in O₂-saturated 0.1 M aqueous KOH electrolyte. (c) CV and (d) LSVs of MnO/NG (different nitrogen wt% annealed at 900 °C) recorded in O₂-saturated 0.1 M aqueous KOH electrolyte. Insets are the corresponding CVs recorded in N₂-saturated 0.1 M aqueous KOH electrolyte.

Table 1 Comparative electrochemical parameters of commercial Pt/C and MnO/N-G annealed at different temperatures (800 °C to 1000 °C) with N contents

S. no.	Catalyst	Reduction peak potential (V vs. RHE)	Reduction peak current density (mA cm^−2)	On-set potential for ORR (V vs. RHE)	Half-wave potential (V vs. RHE)
1	MnO/NG-800	0.61	3.13	0.86	0.74
2	MnO/NG-900 (10%)	0.65	4.13	0.89	0.77
3	MnO/NG-1000	0.66	2.21	0.85	0.71
4	MnO/NG-900 (5%)	0.61	2.69	0.86	0.70
5	MnO/NG-900 (15%)	0.63	4.17	0.89	0.76
6	MnO/G-900	—	—	0.86	0.76
7	Mn3O4/NG-700	—	—	0.82	0.74
8	Commercial Pt/C	—	—	1.01	0.89

---

XPS spectral analysis

To elucidate the electronic state and chemical nature of MnO/NG-900, XPS measurements are performed; the results are shown in Fig. 4. The XPS survey spectrum of the hybrid catalyst (Fig. 4a) indicates the presence of C, Mn, O and N in the catalyst. High resolution XPS spectra of Mn 2p of MnO/NG-900 and Mn₃O₄/G-700 electrocatalysts are shown in Fig. 4b. The peaks observed at 654.1 and 641.3 eV correspond to 2p_1/2 and 2p_3/2, respectively, for MnO/NG-900 catalyst. For this catalyst, the energy separation between 2p_1/2 and 2p_3/2 peaks is calculated and is found to be 12.7 eV, while the energy separation between the peaks for Mn₃O₄/G-700 was found to be 11.8 eV. This higher energy gap observed for MnO/NG-900 is an indication of the transformation of the crystal phase from Mn₃O₄ to MnO. In addition, there is a slight shift in the 2p_3/2 peak to a lower binding energy for MnO/NG-900, which is attributed to the phase change and the interaction with the N-doped support. Fig. 4c shows the de-convoluted XPS spectrum of N for MnO/NG-900, which reveals pyridinic, pyrrolic and graphitic nitrogen species at the binding energies of 398.4, 399.9 and 402.8 eV, respectively.

Fig. 4 (a) XPS survey spectra of graphene and MnO/NG-900 (10%) hybrid catalyst, (b) high resolution Mn 2p spectra of MnO/NG-900 (10%) and Mn₃O₄/NG-700, and (c) de-convoluted N 1s spectra of MnO/NG-900 (10%).

ORR activity evaluation

Fig. 5a shows the comparative LSVs of MnO/NG-900 (10%) and the state-of-art Pt/C catalyst under identical conditions. It is noted that the half-wave potential for Pt/C catalyst is 120 mV higher than that of the MnO/NG-900 (10%) hybrid catalyst. Steady state limiting current densities of around 3.4 and 4.3 mA cm⁻² are obtained for the MnO/NG-900 (10%) and Pt/C catalysts, respectively. To investigate the kinetic behaviour of the electrocatalysts, the Tafel slopes are determined by considering the kinetic current densities of the electrocatalysts. The kinetic current densities are calculated from the mass-transport correction of RDE by the following equation:⁴³where j_k is the mass-transport corrected kinetic current density, j_L is the determined limiting current density and j is the measured current density. The j_k values are then plotted against the potential corresponding to the kinetic control region, as shown in Fig. 5b. The Tafel slopes for the MnO/NG-900 (10%) and Pt/C catalysts are measured to be 81 and 94 mV dec⁻¹, respectively. The comparative electro-kinetic parameters are given in Table 2.

Fig. 5 Comparative (a) LSVs and (b) Tafel plots of MnO/NG-900 (10%) and commercial Pt/C, (c) LSVs at different rotational rates and (d) corresponding K–L plots at different potentials of MnO/NG-900 (10%). (e) and (f) show the comparative ring currents of the RRDE voltammograms of MnO/NG-900 (10%) and commercial Pt/C in O₂-saturated 0.1 M aqueous KOH electrolyte at a rotational rate of 1600 rpm and the calculated electron transfer number and HO₂⁻ generated during the ORR process, respectively.

Table 2 Comparative kinetic parameters, including kinetic current density (i_k), Tafel slope (b), and number of electrons transferred during the ORR process, calculated through K–L plots and RRDE measurements, and percentage peroxide yield

S. no.	Catalyst	ik@0.85 V (mA cm^−2)	Tafel slope, b (mV dec^−1)	Average no. of electrons transferred from K–L plot	Average no. of electrons transferred from RRDE	Percentage peroxide yield (%)
1	MnO/NG-900 (10%)	0.45	81	3.7	3.6	19
2	Pt/C (40%)	11.2	94	Not shown	3.99	0.02

---

The electro-kinetics of MnO/NG-900 (10%) is further examined by recording hydrodynamic voltammograms at different rotations (400, 800, 1200, 1600 and 2000 rpm); the results are shown in Fig. 5c. For this catalyst, the limiting current densities increased with respect to the rotational speed of the electrode, which is attributed to the shortened O₂ diffusion distance between the electrode and the electrolyte interface. The Koutecky–Levich (K–L) plots for MnO/NG-900 (10%) at various potentials are shown in Fig. 5d, using the following kinetic equations:⁴³

where j is the measured current density, j_k is the kinetic current density, j_L is the diffusion-limiting current density, ω is the angular velocity (rad s⁻¹) and B is a parameter calculated from the following equation:

B = 0.62nFC_O2(D_O2)^2/3ν^−1/6 (4)

where n is the number of electrons transferred per O₂ molecule, F is the Faraday constant (96 485 C mol⁻¹), C_O2 is the concentration of oxygen in the electrolyte (1.2 × 10⁻⁶ mol cm⁻³), D_O2 is the diffusion co-efficient of oxygen in the solution (1.9 × 10⁻⁵ cm² s⁻¹) and ν is the kinematic viscosity of 0.1 M aqueous KOH electrolyte (0.01 cm² s⁻¹). The linearity and parallelism of the K–L plots correspond to first order reaction kinetics with regard to the concentration of dissolved O₂. The number of electrons transferred (n) is calculated from the slopes of the K–L plots to be ∼3.7 in the potential range between 0.45 and 0.65 V. It is noteworthy that the hybrid catalyses ORR through a favourable pathway wherein OH⁻ is formed as a dominant byproduct. The enhanced electrocatalytic activity essentially originates from the synergistic effect of both manganese oxide and the active sites of N-doped graphene. This effect can be explained by the presence of nitrogen situated at the interface of the metal oxide and the graphene layers in the hybrid material. This synergistic effect has been explained by Olson et al. through a dual-site mechanism in a Co/polypyrrole system, wherein molecular oxygen is reduced to HO₂⁻ at Co–N–C sites and further reduced to OH⁻ at Co_xO_y/Co active sites.⁴⁴ Tan et al. also described the effects of synergism in MnO–N–C electrocatalyst.⁴⁵ Hence, the presence of nitrogen at the interface actually plays a vital role in the ORR in aiding the electron transfer process. To study the ORR process without nitrogen in the hybrid catalyst, LSV was also carried out, as shown in Fig. S3.† It is noteworthy that MnO/G-900 shows a half-wave potential of 0.76 V, which is 10 mV less than that of MnO/NG-900 (10%). On the other hand, the ORR activity of Mn₃O₄/G-700 was also evaluated, and the corresponding LSV is shown in Fig. S3.† The onset and the half-wave potentials were found to be 0.82 and 0.74 V, respectively. To gain more insights into the electrocatalytic activity of the hybrid catalyst, LSVs were recorded for bare MnO (un-supported) and N-graphene (Fig. S3†). It can be clearly seen that bare MnO shows an inconclusive half-wave response and for N-graphene, the onset and half-wave potentials are found to be 0.74 and 0.63 V, respectively. In the case of MnO/NG-900 (10%) hybrid catalyst, the onset potential is increased (0.89 V vs. RHE), which is due to the synergistic effects of MnO and N-graphene. Moreover, the MnO surface provides active sites for the adsorption of O species followed by electrocatalytic reduction to produce HO₂⁻. Further, the synergistic effect of MnO/NG-900 (10%) greatly reduces the adsorption energy of O species on MnO and hence increases the electrocatalytic ORR. The developed catalyst in this study is compared with other non-precious metal catalysts; electrochemical parameters such as the onset potentials for O₂ reduction, half wave (E_1/2) potentials, and Tafel slopes are provided in Table S1 of the ESI.† From the table, it can be seen that the MnO/N-G-900 catalyst developed in this study shows greatly improved electrochemical behaviour compared to other catalysts.

To further verify the ORR pathway of MnO/NG-900 (10%) catalyst, rotating ring-disk electrode (RRDE) measurements are performed and the formation of peroxide species (HO₂⁻) is monitored throughout the process. The percentage peroxide yield and the number of electrons transferred (n) are calculated using the following equations.⁴³

where I_d is the disk current, I_r is the ring current and N is the current collection efficiency of the Pt ring. Fig. 5e shows the ring current of MnO/NG-900 (10%) compared with the benchmarked Pt/C (40 wt%) catalyst. The measured number of electrons transferred during the ORR process is found to be 3.6 for MnO/NG-900 (10%), which is in good agreement with the observation from the K–L plot. The state-of-the-art Pt/C catalyst shows 3.99 electrons transferred during the ORR process. In addition to the electron transfer number, the percentage peroxide yields are also calculated and are found to be 19% and 0.02% for MnO/NG-900 (10%) and state-of-the-art Pt/C, respectively. The electron transfer number and percentage peroxide yield are calculated throughout the potential range from 0.3 to 0.8 V, as shown in Fig. 5f.

Furthermore, the durability of the electrocatalyst is one of the major obstacles to the long-term operations of fuel cells. In the present study, the optimised MnO/NG-900 (10%) electrocatalyst is subjected to 10 000 continuous potential cycles, as shown in Fig. 6a. It is interesting to see that even after 10 000 potential cycles, the MnO/NG-900 catalyst still retained its ORR characteristics, with well-defined kinetic and diffusion-controlled regions, with a ∼50 mV negative shift in its half-wave potential (E_1/2), and a 20% decrease in limiting current density. However, under similar operating conditions, the commercial Pt/C showed a very large negative shift from the initial cycle and undergoes several redox steps, thereby making inconclusive decision regarding the shift of the E_1/2 potential (Fig. 6a inset). Moreover, after potential cycling, Pt/C showed a clear two-step curve, indicating significant peroxide formation and the loss of the initial 4e⁻ reduction pathway on Pt/C. This could be due to Pt particle agglomeration along with carbon corrosion, which results in loss of the electrochemical active surface area and ORR activity. This high instability of the Pt/C catalyst was also reported by A. Zadick et al.³⁶ From the above, it is clear that the MnO/NG-900 catalyst shows excellent durability in comparison with Pt/C catalyst. The better durability of MnO/NG-900 catalyst is attributed to the stable graphene support and N doping, which further enhances the metal-support interaction and acts as tapping states to prevent detachment of the metal nanoparticles.⁴⁶

Fig. 6 (a) LSVs of MnO/NG-900 (10%) hybrid catalyst before and after a durability test up to 10 000 potential cycles recorded in O₂-saturated 0.1 M aqueous KOH electrolyte with a rotational rate of 1600 rpm. Inset is the LSVs of commercial Pt/C before and after a durability test up to 5000 potential cycles. (b) Polarization and power density curves of AEMFCs comprising MnO/NG-900 (10%) as a cathode catalyst in H₂–O₂ feeds at ambient pressure and temperature (inset: commercial Pt/C as the cathode catalyst). The fuel and oxidant flows were maintained at 200 mL min⁻¹ by mass flow controllers.

Fuel cell performance evaluation

After various physical characterization and electrochemical studies, the MnO/NG-900 (10%) hybrid catalyst exhibited excellent performance with enhanced catalytic activity in the present study. The steady state fuel cell polarization and performance curves for MEA with MnO/NG-900 (10%) as the cathode catalyst are shown in Fig. 6b. The cathode catalyst loadings are maintained to 2 mg cm⁻², and commercial Pt/C is used as anode catalyst with a loading of 0.5 mg cm⁻². The performance is compared with commercial Pt/C as the cathode catalyst under identical operating conditions. Commercial Pt/C is used as anode catalyst for all the MEAs; therefore, any variation in the polarisation performance can only be attributed to the ORR behaviour of the hybrid catalysts. Remarkably, the AEMFC with MnO/NG-900 (10%) catalyst delivered a peak power density of about 13 mW cm⁻² at a load current density of 32.5 mA cm⁻² under ambient pressure and temperature. The Pt/C catalyst showed a peak power density of about 69 mW cm⁻² at a load current density of 150 mA cm⁻². Although the power density of MnO/NG-900 catalyst is significantly lower (∼6 times) than that of the commercial Pt/C catalyst, this is appreciable as a non-precious metal catalyst, which opens up a new path to develop novel and cost-effective electrocatalysts for the oxygen reduction reaction for fuel cell technology. Moreover, the MnO/NG-900 catalyst was found to be more durable than the Pt/C catalyst.

Conclusions

A MnO/NG hybrid catalyst was successfully synthesized as a durable, non-precious metal catalyst for ORR in alkaline medium, and its potential application in AEMFCs is demonstrated for the first time in the literature. After systematically optimizing the reaction temperature and N content in graphene, it was found that MnO/NG-900 (10%) catalyst demonstrates the highest ORR catalytic activity, with a dominant 4e⁻ transfer process that was confirmed by both RDE and RRDE measurements. In comparison with commercial Pt/C catalyst, MnO/NG-900 (10%) catalyst has less ORR activity in alkaline medium but much better durability, as observed from LSV data after 10 000 repeated potential cycles. The optimized MnO/NG-900 (10%) catalyst was used as the cathode catalyst in a single cell AEMFC; a peak power density of 13 mW cm⁻² at 30 °C was observed, which appears favorable for the development of a durable and non-precious metal catalyst. This provides new insights and understanding for the synthesis of more advanced non-precious metal-oxide based catalysts for the efficient reduction of O₂, which is useful in alkaline polymer electrolyte fuel cells.

Acknowledgements

Financial support from CSIR, New Delhi, India, through HYDEN (CSC 0122) is gratefully acknowledged. The authors thank Dr Vijayamohanan K. Pillai, director, CSIR-CECRI and Dr V. V. Giridhar, scientist-in-charge, CSIR-CECRI, Madras unit for their constant encouragement and support. M. V. thanks Dr A. Karthigeyan, department of physics and nanotechnology, SRM University, for his support.

References

1. A. S. Aricò, P. Bruce, B. Scrosati, J. M. Tarascon and W. Schalkwijk, Nat. Mater., 2005, 4, 366.
2. J. M. Andújar and F. Segura, Renewable Sustainable Energy Rev., 2009, 13, 2309.
3. A. A. Gewirth and M. S. Thorum, Inorg. Chem., 2010, 49, 3557.
4. J. M. Tang, K. Jensen, M. Waje, W. Li, P. Larsen, K. Pauley, Z. Chen, P. Ramesh, M. E. Itkis, Y. Yan and R. C. Haddon, J. Phys. Chem. C, 2007, 111, 17901.
5. A. Morozan, B. Jousselme and S. Palacin, Energy Environ. Sci., 2011, 4, 1238.
6. S. G. Peera, A. K. Sahu, A. Arunchander, K. Nath and S. D. Bhat, J. Power Sources, 2015, 297, 379.
7. A. Arunchander, S. G. Peera, V. Parthiban, S. Akula, K. K. Tintula, S. D. Bhat and A. K. Sahu, RSC Adv., 2015, 5, 75218.
8. K. K. Tintula, A. K. Sahu, S. D. Bhat, S. Pitchumani and P. Sridhar, Appl. Catal., B, 2011, 110, 178.
9. S. G. Peera, K. K. Tintula, A. K. Sahu, S. Shanmugam, P. Sridhar and S. Pitchumani, Electrochim. Acta, 2013, 10, 95.
10. K. K. Tintula, A. K. Sahu, A. Shahid, S. Pitchumani, P. Sridhar and A. K. Shukla, J. Electrochem. Soc., 2011, 158, B622.
11. G. Merle, M. Wessling and K. Nijmeijer, J. Membr. Sci., 2011, 377, 1.
12. C. Zhang, N. Mahmood, H. Yin, F. Liu and Y. Hou, Adv. Mater., 2013, 25, 4932.
13. Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, X. Chen and S. Huang, ACS Nano, 2012, 6, 205.
14. Z. H. Sheng, H. L. Gao, W. J. Bao, F. B. Wang and X. H. Xia, J. Mater. Chem., 2012, 22, 390.
15. T. V. Vineesh, M. Praveen Kumar, C. Takahashi, G. Kalita, S. Alwarappan, D. K. Pattanayak and T. N. Narayanan, Adv. Energy Mater., 2015, 5, 1500658.
16. S. A. Wohlgemuth, R. J. White, M. G. Willinger, M. M. Titirici and M. Antonietti, Green Chem., 2012, 14, 1515.
17. H. Jiang, Y. Zhu, Q. Feng, Y. Su, X. Yang and C. Li, Chem.–Eur. J., 2014, 20, 3106.
18. S. G. Peera, A. K. Sahu, A. Arunchander, S. D. Bhat, J. Karthikeyan and P. Murugan, Carbon, 2015, 93, 130.
19. S. Gupta, D. Tryk, I. Bae, W. Aldred and E. Yeager, J. Appl. Electrochem., 1989, 19, 19.
20. T. Odedairo, X. Yan, J. Ma, Y. Jiao, X. Yao, A. Du and Z. Zhu, ACS Appl. Mater. Interfaces, 2015, 7, 21373.
21. S. Guo, S. Zhang, L. Wu and S. Sun, Angew. Chem., Int. Ed., 2012, 51, 11770.
22. Z. S. Wu, S. Yang, Y. Sun, K. Parvez, X. Feng and K. Müllen, J. Am. Chem. Soc., 2012, 134, 9082.
23. J. Zhang, C. Guo, L. Zhang and C. M. Li, Chem. Commun., 2013, 49, 6334.
24. N. Ominde, N. Bartlett, X. Q. Yang and D. Qu, J. Power Sources, 2010, 195, 3984.
25. S. Cao, N. Han, J. Han, Y. Hu, L. Fan, C. Zhou and R. Guo, ACS Appl. Mater. Interfaces, 2016, 8, 6040.
26. K. H. Wu, Q. Zeng, B. Zhang, X. Leng, D. S. Su, I. R. Gentle and D. W. Wang, ChemSusChem, 2015, 8, 3331.
27. S. Bag, K. Roy, C. S. Gopinath and C. R. Raj, ACS Appl. Mater. Interfaces, 2014, 6, 2692.
28. M. S. El-Deab and T. Ohsaka, Angew. Chem., Int. Ed., 2006, 45, 5963.
29. I. Roche, E. Chaînet, M. Chatenet and J. Vondrák, J. Phys. Chem. C, 2007, 111, 1434.
30. J. Suntivich, H. A. Gasteiger, N. Yabuuchi, H. Nakanishi, J. B. Goodenough and Y. Shao-Horn, Nat. Chem., 2011, 3, 546.
31. F. Cheng, J. Shen, B. Peng, Y. Pan, Z. Tao and J. Chen, Nat. Chem., 2011, 3, 79.
32. K. L. Pickrahn, S. W. Park, Y. Gorlin, H. B. R. Lee, T. F. Jaramillo and S. F. Bent, Adv. Energy Mater., 2012, 2, 1269.
33. L. Mao, D. Zhang, T. Sotomur, K. Nakatsu, N. Koshib and T. Ohsaka, Electrochim. Acta, 2003, 48, 1015.
34. B. Klápště, J. Vondrák and J. Velická, Electrochim. Acta, 2002, 47, 2365.
35. R. Chen, J. Yan, Y. Liu and J. Li, J. Phys. Chem. C, 2015, 119, 8032.
36. A. Zadick, L. Dubau, N. Sergent, G. Berthomé and M. Chatenet, ACS Catal., 2015, 5, 4819.
37. K. Hari Gopi, S. G. Peera, S. D. Bhat, P. Sridhar and S. Pitchumani, Int. J. Hydrogen Energy, 2014, 39, 2659.
38. A. Gil, L. M. Gandía and S. A. Korili, Appl. Catal., A, 2004, 274, 229.
39. X. Wang and Y. C. Xie, Catal. Lett., 2001, 72, 51.
40. E. R. Stobbe, B. A. de Boer and J. W. Geus, Catal. Today, 1999, 47, 161.
41. A. Pozio, M. D. Francesco, A. Cemmi, F. Cardellini and L. Giorgi, J. Power Sources, 2002, 105, 13.
42. F. H. B. Lima, M. L. Calegaro and E. A. Ticianelli, Electrochim. Acta, 2007, 52, 3732.
43. C. Song and J. Zhang, in PEM Fuel Cell Electrocatalysts and Catalyst Layers: Fundamentals and Applications, ed. J. Zhang, Springer-Verlag, London, 2008, ch. 2, p. 89.
44. T. S. Olson, S. Pylypenko, P. Atanassov, K. Asazawa, K. Yamada and H. Tanaka, J. Phys. Chem. C, 2010, 114, 5049.
45. Y. Tan, C. Xu, G. Chen, X. Fang, N. Zheng and Q. Xie, Adv. Funct. Mater., 2012, 22, 4584.
46. S. G. Peera, A. Arunchander and A. K. Sahu, Carbon, 2016, 107, 667.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra20627a

---