Electrocatalysis of both oxygen reduction and water oxidation using a cost-effective three-dimensional MnO₂/graphene/carbon nanotube

Abstract

The electrochemical oxygen reduction reaction (ORR) and the oxygen evolution reaction (OER) are of great interest since they are involved in energy conversion between fuel and electricity. Here, we developed a bifunctional MnO₂/graphene/carbon nanotube that is free of noble metals and that could be a promising candidate electrocatalyst for these oxygen reduction and evolution reactions. It was shown to be able to act as an efficient cathode catalyst for the ORR, having a positive half-wave potential that differs by only ∼55 mV from that of commercial Pt/C, and a high cathodic current density that is comparable to that of the Pt/C catalyst. Moreover, the hybrid exhibited superior durability with nearly no decay in ORR activity even after 10 000 s of continuous operation in 0.1 M KOH, while Pt/C shows a 20% decrease in the activity. Most importantly, the hybrid was also shown to be highly active for the OER. These observations show this hybrid to be a high-performance non-precious metal-based bi-catalyst for both the ORR and OER.

---

1. Introduction

Oxygen electrochemistry has attracted growing interest because the conversion between O₂ and H₂O plays important roles in renewable energy technologies. The oxygen reduction reaction (ORR) is the ubiquitous cathode reaction in fuel cells, while the oxygen evolution reaction (OER) is the anode reaction employed in the electrolysis cell.¹ Both the fuel cell reaction and the water electrolysis reaction require large over-potentials at the oxygen electrode – and no current catalyst material operates at or near the equilibrium potential for either the ORR or the OER.² The ORR takes place almost exclusively on Pt/C catalysts at the cathode in lithium-air batteries and low-temperature fuel cells. However, the high price, sluggish ORR process, intolerance to fuel crossover, and instability of the Pt/C catalyst in the fuel cell environment have greatly impeded its commercialization and limited its performance.^3,4 At the same time, Pt has only moderate activity for the OER. Ruthenium (Ru) and iridium (Ir) oxides are OER catalysts with moderate over-potentials, but they still have some shortcomings. Thus, the development of a bifunctional catalyst with significant overpotentials for both reactions and that excludes precious metals, which are both scarce and expensive, remains a major challenge. It is therefore very urgent to develop efficient bifunctional catalysts, i.e., for both ORR and OER, particularly for unitized regenerative fuel cells, which can form a promising energy storage system that works both as a fuel cell and in reverse as a water electrolyzer that produces H₂ and O₂ to feed the fuel cell.

For this purpose, we have applied a facile one-pot reaction process to synthesize MnO₂/graphene/carbon nanotubes as efficient bifunctional oxygen catalysts. Manganese oxides have been widely investigated for their prominent advantages, which arise from their low cost and toxicity, as well as their high chemical stability and catalytic activity.^5–9 More importantly, molecular mimics of manganese oxide surfaces indicated it is a potentially interesting material for bifunctional oxygen catalysis.¹ However, the poor conductivity of MnO₂ (∼1 × 10⁵ S cm⁻¹) limits its applications. In order to enhance the electrical conductivity of MnO₂ electrodes, tremendous efforts have been focused on combining highly conductive materials such as graphene, carbon nanotubes (CNTs), porous carbon, and activated carbons.^10–12 We produced a hybrid carbon matrix, consisting of a CNT matrix and graphene partly exfoliated from the walls of the CNT matrix, that not only provides a large surface area but also greatly improves the conductivity due to the role of CNT in bridging graphene. Meanwhile, the MnO₂ nanorods strongly coupled with the graphene to exert the synergistic effects between MnO₂ and graphene, and also MnO₂ can escape from the detachment, aggregation and even dissolution with the graphene during potential cycling. This strongly interaction was extremely beneficial for the stability of electrocatalytic reduction of oxygen. Therefore, the MnO₂/graphene/carbon nanotubes that we fabricated exhibited surprisingly high performance in both the ORR and OER, with a catalytic activity even comparable to that of a commercial carbon-supported Pt catalyst in alkaline solutions. Simultaneously, the hybrid also showed superior stability, thus giving rise to a new bi-functional catalyst for ORR and OER.

2. Experimental

2.1 Apparatus and reagents

All experiments were performed in 0.1 M KOH, and doubly distilled water was used throughout.

Electrochemical experiments including cyclic voltammetry (CV), linear sweep voltammetry (LSV) and amperometry were carried out on a CHI 660C electrochemical workstation (Chenhua, Shanghai, China). A conventional three-electrode electrochemical system was used for all electrochemical experiments, which consisted of a working electrode, a Pt foil auxiliary electrode and a saturated calomel electrode. A glassy carbon electrode (GCE) was used as the basal working electrode. Transmission electron microscopy (TEM) images were taken with a JEOL 2011 microscope (Japan) operated at 200 kV. Raman spectra were recorded on a LabRam-1B microscopic confocal Raman spectrometer (Jobin Yvon, France). Powder X-ray diffraction (XRD) measurements were performed on a Bruker D4 diffractometer at a scanning rate of 1° min⁻¹ in the 2θ range from 10 to 80° (Cu Kα radiation, λ = 1.54056 Å). Thermogravimetric (TG) analysis of the sample was performed on a Pyris Diamond TG/DTA thermo-gravimetric analyzer (Perkin-Elmer Thermal Analysis).

2.2 Preparation of MnO₂/G/CNT hybrid

The MnO₂/G/CNT hybrid was prepared using a one-pot method. In a typical procedure, MWCNTs (10 mg) were dispersed in 10 mL of water including 100 μL of H₂SO₄ and 400 μL of HNO₃ under stirring at 70 °C for 30 min. Afterwards, 300 μL of H₂SO₄ was added into the above mixture and kept at 0 °C for 1 h. Then, permanganate crystallites (50 mg) were added and stirred at 55 °C for 2 h. The reaction mixture was then removed from the heat source, and cooled and diluted by being poured into 500 mL of doubly distilled water. The solution was filtered over a polytetrafluoroethylene membrane (0.22 μm pore size, Aldrich), and the remaining solid was washed repeatedly with distilled water several times. The obtained products were called MnO₂/G/CNT.

C/MnO₂ was synthetized using graphite and permanganate, and the specific process was the same as the procedure used for the synthesis of MnO₂/G/CNT.

2.3 Preparation of modified electrodes and electrochemical test

Prior to use, a glassy carbon rotating disc electrode (GCRDE) was polished with 0.3 and 0.05 μm α-Al₂O₃ powder until a mirror-shiny surface was obtained, and then ultrasonicated in ethanol and doubly distilled water for 10 min, respectively. Finally, it was dried in a stream of high-purity nitrogen for further use. 2 mg of the MnO₂/G/CNT samples were dispersed in a solution containing 1 mL of deionized water and 0.5 mL of 1 wt% Nafion aqueous solution. The mixture was ultrasonicated to obtain a homogenous catalyst ink. Then 10 μL of the resulting mixture was dropped onto the cleaned GCRDE surface to prepare MnO₂/G/CNT/GCE, C/MnO₂/GCRDE, or CNT/GCRDE and the modified electrode was allowed to dry under an infrared lamp for 10 min. The obtained modified electrodes were preserved in a refrigerator at 4 °C after being washed with doubly distilled water.

Before testing the electrode, an O₂-saturated electrolyte solution and an N₂-saturated electrolyte solution were made by having one gas or the other flow through the electrolyte in the cell for 20 min. The cell was kept in a 25 °C water bath for all the electrochemical tests. Linear sweep voltammograms of the modified GCRDE were recorded in O₂-saturated 0.1 M KOH with a scan rate of 5 mV s⁻¹ at various rotating speeds from 225 to 1600 rpm.

3. Results and discussion

3.1 Characterization of the MnO₂/G/CNT hybrid

Fig. 1A, Fig. 1SA and SB† present the TEM images of the MnO₂/G/CNT, which display a long-range array of transparent petal-like sheets. High-resolution TEM image (Fig. 1SA†) showed that the nanorod were closely attached to graphene. Fig. 1B shows SEM images of the hybrid, which have a reticular morphology and curved petal-like walls on the carbon nanotubes over the entire length. To gain further insight into the structure of the hybrid, excess permanganate and MnO₂ were removed. Sufficient hydrogen peroxide and 1 M HCl solution were added to the samples, which were then washed with a large amount of distilled water repeatedly. Fig. S2A† shows a TEM image of the samples after the removal of MnO₂, revealing that the tidy petal-like structures collapsed without the support of MnO₂ but that the stripped graphene remained around the CNT. We used the Raman technique to investigate the vibrational properties of the treated samples (Fig. S2B†). Compared to the samples without treatment, the Raman result for the treated sample shows that the obvious characteristic peaks located at 640 cm⁻¹ disappeared, demonstrating that many MnO₂ nanocrystals were destroyed after treatment. Raman spectra of the MnO₂/G/CNT and CNT are displayed in Fig. 1C. Both the CNT and MnO₂/G/CNT display two prominent peaks at 1573 and 1325 cm⁻¹, which correspond to the well-documented G and D bands, respectively. The ratio of the intensity of the D band to that of the G band (I_D/I_G) increased accordingly from 0.13 to 0.38 as the oxidation reaction progressed, indicating the enhanced level of disorder of the product and the formation of sp³ carbon after functionalization. The Raman band at 640 cm⁻¹ can be recognized as the symmetric stretching vibration (Mn–O) of the MnO₆ groups,¹³ which can be used to characterize the tunnel species of α-type MnO₂ materials. The XRD patterns of the MnO₂/G/CNT and pristine CNT are shown in Fig. 1D. The XRD pattern of MnO₂/G/CNT shows reflections at 28.82°, 37.44°, 41.94°, 49.83°, 56.39°, 60.27°, and 69.69°, which are characteristic of the standard XRD pattern of MnO₂ (JCPDS 44-0141), as well as a reflection at 26.33°, which is ascribed to the crystal plane of graphite and is consistent with the XRD pattern of CNT (Fig. 1D). The results indicate that the final product is composed of crystalline MnO₂ and CNT. Typical XPS spectra for the MnO₂/G/CNT are shown in Fig. 1E and F. The peaks of Mn (2p_3/2, 2p_1/2), O 1s and C 1s can be observed in the survey spectrum. Furthermore, the Mn 2p_3/2 peak is centered at 643.2 eV and the Mn 2p_1/2 peak is located at 654.6 eV, with a spin-orbit splitting of 11.4 eV, suggesting that the predominant manganese oxidation state is Mn(IV). These results are in accordance with the previous reports.¹⁴

Fig. 1 Characterization of the MnO₂/G/CNT hybrid: (A) TEM image of the hybrid; (B) SEM image of the hybrid; (C) Raman spectra of the hybrid and of CNT; (D) XRD patterns of the hybrid and of CNT; (E) XPS survey spectrum of MnO₂; (F) Mn 2p XPS spectrum of MnO₂.

Fig. 2A shows a TEM image of a representative segment of the hybrid. Fig. 3B–D shows elemental distribution of Mn, O and C in the hybrid. The mapping results indicate that the distributions of the C, Mn and O elements are homogeneous throughout the petal-like nanosheets, implying that the nanosheets are composed of manganese oxide and C. To gain additional information about the relative content of MnO₂ in the hybrid, a thermogravimetric (TG) investigation was carried out in air and the curve is shown in Fig. 3. The content of MnO₂ in the hybrid was 52.9 wt% according to the TG curve.

Fig. 2 (A) TEM image of a representative segment of the hybrid; (B–D) EDX mapping results for Mn, O and C elements.

Fig. 3 TG analysis of MnO₂/G/CNT.

3.2 Electrocatalytic properties of the hybrid for the ORR

First, to gain insight into the ORR activity of MnO₂/G/CNT/GCRDE, we examined the electrocatalytic properties of MnO₂/G/CNT/GCRDE in N₂- and O₂-saturated 0.1 M aqueous KOH electrolyte solutions, using cyclic voltammetry at a scan rate of 100 mV s⁻¹ (Fig. 4A). A quasi-rectangular voltammogram without an obvious redox peak was obtained when the electrolyte was saturated with N₂. In contrast, when the electrolyte was saturated with O₂, the MnO₂/G/CNT hybrid showed a well-defined characteristic ORR cathodic peak at 0.7 V (vs. RHE), suggesting that the GCRDE modified by the MnO₂/G/CNT hybrid displayed pronounced electrocatalytic activity for oxygen reduction. For the purpose of comparison, the LSV curves for the commercial catalyst (Pt/C, 20 wt%), MnO₂/G/CNT, C/MnO₂, and CNT are shown in Fig. 4B. The same amount of each catalyst by mass (0.068 mg cm⁻²) was loaded onto each GCRDE. The onset potential of the MnO₂/G/CNT hybrid electrode towards the ORR was observed to be close to that of Pt/C, whereas the C/MnO₂ and CNT electrodes commenced oxygen reduction at more negative potentials. Moreover, the plateau limiting current of MnO₂/G/CNT (3.76 mA cm⁻²) is much larger than that of CNT and than that of C/MnO₂, and is even comparable to that of Pt/C, which may be due to MnO₂/G/CNT having a larger electroactive surface area and the synergistic effect generated from MnO₂, G, and CNT. Meanwhile, manganese oxide itself is known to have considerable activity for the oxygen reduction and oxygen evolution reactions.^15,16 Also, the structure of G/CNT favours the electron transfer and thus boosts ORR activity.

Fig. 4 (A) CV of a GCE modified by the MnO₂/G/CNT hybrid, in 0.1 M KOH with O₂ (red) and N₂ (black) saturation; (B) LSVs of different materials in O₂-saturated 0.1 M KOH with a sweep rate of 5 mV s⁻¹ at a rotation rate of 1600 rpm.

3.3 Kinetics, stability and anti-poisoning properties

To obtain additional insight about the ORR process on the MnO₂/G/CNT hybrid, LSVs on GCRDE were recorded at different rotating speeds from 0 rpm to 2000 rpm in 0.1 M KOH electrolyte saturated with O₂ (Fig. 5A). In Fig. 5B, a good linear fit can be seen for each potential with identical slopes. For the oxygen reduction on a GCRDE, Koutecky–Levich (K–L) plots were analyzed at different electrode potentials. The kinetic parameters can be analyzed on the basis of the K–L equations.

J⁻¹ = J⁻¹_L + J⁻¹_K (1)

J_L = 0.62nFC₀D^2/3v^−1/6w^1/2 (2)

J_K = nFkC₀ (3)

Where J is the current density, J_K and J_L are the kinetic- and diffusion-limiting current densities, w is the angular velocity of the disk (w = 2πN, N being the rotation frequency), n is the overall number of electrons transferred upon oxygen reduction, F is the Faraday constant (F = 96 485 C mol⁻¹), D is the diffusion coefficient of O₂ in 0.1 M KOH electrolyte (1.9 × 10⁻⁵ cm² s⁻¹), C₀ is the bulk concentration of O₂ (1.2 × 10⁻³ mol L⁻¹), v is the kinetic viscosity of the electrolyte (0.01 cm² s⁻¹) and k is the electron-transfer rate constant.

Fig. 5 (A) LSVs of MnO₂/G/CNT at different rotation rates in 0.1 M KOH solutions saturated with O₂ (B) K–L plots of J⁻¹ vs. ω^1/2 obtained from the LSV data at different potentials.

The slopes of ω^−1/2 vs. J⁻¹ were used to calculate the number of electrons transferred (n) in ORR. This number was calculated to be 3.95, 4.00, 3.98, and 3.96 at 0.8, 0.75, 0.7, 0.65 and 0.6 V, respectively, suggesting that the MnO₂/G/CNT hybrid favours a 4e oxygen reduction process, similar to the ORR catalysed by a high-quality commercial Pt/C catalyst measured in the same 0.1 M KOH electrolyte (n ∼ 4 for Pt/C) (Fig. 5C and D). So, by taking into consideration the work reported by Roche et al.,¹⁶ we propose that the mechanism for the ORR may be as follows:

MnO₂ + H₂O + e⁻ ↔ MnOOH + OH⁻ (1)

2MnOOH + O₂ ↔ (MnOOH)₂⋯O_2,ads (2)

(MnOOH)₂⋯O_2,ads + e⁻ → MnOOH⋯O_ads + OH⁻ + MnO₂ (3)

MnOOH⋯O_ads + e⁻ ↔ MnO₂ + OH⁻ (4)

(MnOOH)₂⋯O_2,ads represents the adsorption of one oxygen molecule onto two neighboring MnOOH sites. In such a process, the Mn^III/Mn^IV species act as oxygen acceptor/donor.

The RDE measurements for the C/MnO₂ and MWCNT electrocatalyst were also carried out in our experiments. As seen in Fig. S3 in ESI,† C/MnO₂ has an electron transfer number of ∼4. MWCNT also has an ORR response, but its activity is less than desirable. The numbers of electrons transferred per O₂ for MWCNT is calculated to be 2.38, 1.7, and 1.8 at 0.6, 0.5 and 0.4 V, respectively (Fig. S4†). This suggests that oxygen reduction on the MWCNT electrocatalyst may proceed by a coexisting pathway involving both the two-electron and four-electron transfers.

Fig. 6 shows tests of the stability of MnO₂/G/CNT and Pt/C for the ORR. It can be seen that the current for the ORR at the Pt/C electrode decreased by nearly 20% after 40 000 s of continuous operation in 0.1 M KOH (Fig. 6). In contrast, our hybrid exhibited superior durability, with little decay in the ORR activity, giving higher long-term ORR currents than did the Pt/C electrode.

Fig. 6 Chronoamperometric responses of MnO₂/G/CNT- and Pt/C-modified electrodes kept at 0.74 V versus RHE in O₂-saturated 0.1 M KOH.

To examine possible crossover effects, the electrocatalytic selectivity of the MnO₂/G/CNT hybrid was measured against the electro-oxidation of methanol in 0.1 M KOH saturated with O₂ as shown in Fig. 7. No noticeable change was observed in the oxygen-reduction current compared to MnO₂/G/CNT in O₂-saturated 0.1 M KOH without methanol. Thus, the MnO₂/G/CNT hybrid exhibited high selectivity in avoiding crossover effects for the ORR.

Fig. 7 CVs of MnO₂/G/CNT in O₂-saturated 0.1 M KOH and O₂-saturated 0.1 M KOH with 1 M methanol.

3.4 Electrocatalytic properties toward the OER

Finally, we extended the potential of our hybrid electrode to 2.25 V versus RHE for the water oxidation regime and evaluated the electrocatalytic oxygen evolution reaction (OER). Regarding OER activity, the MnO₂/G/CNT hybrid was clearly more active than Pt/C and C/MnO₂. From Fig. 8, MnO₂/G/CNT was found to be highly active for both the ORR and the OER. These results make our hybrid material a powerful bi-functional catalyst for both oxygen reduction and water oxidation.

Fig. 8 Oxygen electrode activities within the ORR and OER potential window of C/MnO₂ and MnO₂/G/CNT in O₂-saturated 0.1 M KOH.

4. Conclusion

To sum up, we fabricated a MnO₂/G/CNT hybrid from inexpensive and earth-abundant materials, and showed that this hybrid is a catalyst that exhibits excellent bifunctional oxygen electrode activity. The special features of the MnO₂/G/CNT hybrid include its large active surface areas that accelerate the interfacial electrochemical reaction, contributing to their potent ORR and OER activities. Furthermore, this catalyst is comparable to fresh commercial Pt/C catalyst with regards to ORR and OER activities in alkaline solution, and far exceeds Pt/C with regards to stability. These results show that OER activities can be further improved by synergistic coupling of nonprecious functional materials. Such improvement is very much required for energy conversion technologies.

Acknowledgements

This work was supported by The National Natural Science Foundation of China (21175029, 21335002) and the Shanghai Leading Academic Discipline Project (B109).

References

1. Y. Gorlin and T. F. Jaramillo, J. Am. Chem. Soc., 2010, 132, 13612–13614.
2. G. Y. Chen, S. R. Bare and T. E. Mallouk, J. Electrochem. Soc., 2002, 149, A1092–A1099.
3. R. Bashyam and P. Zelenay, Nature, 2006, 443, 63–66.
4. M. Lefevre, E. Proietti, F. Jaouen and J. P. Dodelet, Science, 2009, 324, 71–74.
5. G. H. Yu, L. B. Hu, M. Vosgueritchian, H. L. Wang, X. Xie, J. R. McDonough, X. Cui, Y. Cui and Z. N. Bao, Nano Lett., 2011, 11, 2905–2911.
6. F. Y. Cheng, J. Shen, W. Q. Ji, Z. L. Tao and J. Chen, ACS Appl. Mater. Interfaces, 2009, 1, 460–466.
7. L. Trahey, N. K. Karan, M. Chan, J. Lu, Y. Ren, J. Greeley, M. Balasubramanian, A. K. Burrell, L. A. Curtiss and M. M. Thackeray, Adv. Energy Mater., 2013, 3, 75–84.
8. G. Y. Zhu, Z. He, J. Chen, J. Zhao, X. M. Feng, Y. W. Ma, Q. L. Fan, L. H. Wang and W. Huang, Nanoscale, 2014, 6, 1079–1085.
9. Y. Hou, Y. W. Cheng, T. Hobson and J. Liu, Nano Lett., 2010, 10, 2727–2733.
10. G. H. Yu, L. B. Hu, N. A. Liu, H. L. Wang, M. Vosgueritchian, Y. Yang, Y. Cui and Z. A. Bao, Nano Lett., 2011, 11, 4438–4442.
11. Z. J. Fan, J. Yan, T. Wei, L. J. Zhi, G. Q. Ning, T. Y. Li and F. Wei, Adv. Funct. Mater., 2011, 21, 2366–2375.
12. Y. Chen, Y. Zhang, D. Geng, R. Li, H. Hong, J. Chen and X. Sun, Carbon, 2011, 49, 4434–4442.
13. S. Ma, K. Ahn, E. Lee, K. Oh and K. Kim, Carbon, 2007, 45, 375–382.
14. Z. Liu, X. Tan, X. Gao and L. Song, J. Power Sources, 2014, 267, 812–820.
15. J. Feng, Y. Liang, H. Wang, Y. Li, B. Zhang, J. Zhou, J. Wang, T. Regier and H. Dai, nano research, 2012, 5, 718–725.
16. I. Roche, E. Chainet, M. Chatenet and J. Vondrak, journal of physical chemistry C, 2007, 111, 1434–1443.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra14452j

---