Preparation of a Ag–MnO₂/graphene composite for the oxygen reduction reaction in alkaline solution

Abstract

Ag–MnO₂/graphene composite with mesoporous structure is prepared by an immersion-calcination method. The composite composed of Ag (∼6 nm) and MnO₂ (∼4 nm) nanoparticles is distributed evenly on the graphene sheets. The oxygen reduction peak of the Ag–MnO₂/graphene catalyst in a 0.1 M KOH solution is tested at −0.15 V, which is more positive than that of 20% Pt/C (−0.19 V). The onset potential of Ag–MnO₂/graphene (0.068 V) is close to Pt/C (0.079 V). The oxygen reduction reaction limiting current density for the Ag–MnO₂/graphene catalyst is approximately 5.62 mA cm⁻², which is similar to that for Pt/C (5.83 mA cm⁻²). The onset potential of the Ag–MnO₂/graphene is close to that of Pt/C. The introduction of Ag nanoparticles between the MnO₂ nanoparticles on graphene sheets can significantly improve the catalytic activation, because of the increase in conductivity and abundant active sites provided during the ORR process.

---

1. Introduction

The oxygen reduction reaction (ORR), which is the main cathodic reaction for some clean energy technologies, has great significance for metal air batteries and fuel cells.^1,2 Currently, the widely used catalysts for ORR in acidic media are Pt and Pt alloys because of their stability. However, the high cost of Pt materials is one of the obstacles for the commercialization of fuel cells. Meanwhile, some non-Pt electrocatalysts, such as carbon-supported metals and transition metal oxides (Pd/C, Ag–MnO_x/C), have been investigated extensively to catalyze the ORR in alkaline environments. In addition, the kinetics of the ORR in alkaline solution is superior to that in acidic media.^3–6

Among the nanomaterials, silver is highly active toward the ORR in alkaline solution. This condition is because of high electrical conductivities, antibacterial properties, and excellent catalytic properties.⁷ However, Ag nanoparticles are likely to aggregate because of the extremely high surface energy. Thus, Ag nanoparticles are usually supported on conductive supports, such as like graphene and carbon, to enhance the surface area of the active sites.⁸ However, the activity of silver toward the ORR is not satisfactory and must be improved.

In recent years, MnO₂ has been widely investigated as one of the most promising candidates for the ORR in alkaline solution, because this substance is naturally abundance and low cost, and it has high energy density and environmental compatibility.^9–12 However, the active phases of the MnO₂ reported in the literature are still in conflict, and the catalytic performance of MnO₂ nanostructures toward the ORR is highly dependent upon the crystalline phases of the chemical composition. For example, the orders was β-MnO₂ < λ-MnO₂ < γ-MnO₂ < α-MnO₂ ≈ δ-MnO₂ for different crystal forms of MnO₂.¹³ The advantages of silver and manganese oxides have been combined by preparing an Ag–MnO₂/C catalyst.^14,15

In this study, a facile method was developed to prepare Ag–MnO₂/graphene composite. The as-prepared graphene was obtained with the use of an exfoliated graphite oxide through thermal annealing, which is the support of the Ag–MnO₂ catalyst for ORR in alkaline solution. The composite has excellent electrocatalytic performance, because Ag nanoparticles significantly increase in conductivity, surface area and active sites, which are favorable for ORR.

2. Experiment

2.1. Preparation of catalyst

Analytical grade AgNO₃ and 50 wt% Mn(NO)₃·4H₂O were used without further purification. Pt/C (20 wt%) was purchased from Johnson Mattey.

The Ag–MnO₂/graphene composite was synthesized through a simple method: first, graphite oxide (GO) was prepared from natural graphite with a modified Hummers method.¹⁶ The GO powders was reduced through thermal annealing at 400 °C for 60 min in N₂ flow to achieve graphene.¹⁷ 0.025 g Mn(NO)₃ was added into the homogeneous suspension of 0.05 g graphene in 20 mL absolute ethyl alcohol and ultrasonicated for 15 min, followed by adding 0.01 g AgNO₃ and ultrasonicated for 15 min. The solvent was then evaporated under an infrared lamp, and the composite powder was subjected to thermal decomposition at 320 °C for 45 min in a N₂ atmosphere. The sample was then cooled down to room temperature, and the Ag–MnO₂/graphene composite was successfully obtained. MnO₂/graphene and Ag/graphene composites were synthesized through the same process of the above without addition of AgNO₃ and Mn(NO)₃, respectively.

2.2. Physical characterization

The Ag–MnO₂/graphene composite and MnO₂/graphene were characterized through X-ray diffraction (XRD, D/MAX-2500), Raman spectroscopy (In Via Rflex), Transmission electron microscopy (TEM, Tecnai G2 F20), scanning electron microscopy (SEM, S4800) equipped with an energy dispersive X-ray analyzer (EDX), and X-ray photoelectron spectroscopy (XPS, PHI1600).

2.3. Preparation of modified electrodes and electrochemical measurements

All half-cell experiments for the ORR were conducted under the same conditions where Ni-mesh and Hg/HgO were used as counter and reference electrodes, respectively. Catalyst paste was prepared by ultrasonically mixing 6.0 mg of the as-prepared sample with a certain amount of 5 wt% Nafion solution for 30 min to create a homogeneous suspension. The prepared catalytic paste was transferred to the surface of a glassy carbon electrode with a diameters of 3 mm. Finally, the ink was dried under an infrared lamp to form a thin catalyst film on a glass carbon electrode as a working electrode.

The cyclic voltammograms (CVs) were obtained at different scan rates with a rotation rate of 1200 rpm in a 0.1 M KOH solution, saturated with N₂ and O₂, correspondingly. The linear sweep voltammograms (LSVs) were obtained under the process of saturated with O₂ in a 0.1 M KOH solution at different rotation rates with scan rate of 1 mV s⁻¹, A rotating disk electrode (RDE, ATA-1B, China) and an electrochemical workstation (CHI604C, ShangHai Chen Hua, China) were used for to obtain CVs and LSVs.

3. Results and discussion

3.1. Physical characterization of the composite

The Ag–MnO₂/graphene composite as well as graphene, MnO₂/graphene and Ag/graphene were characterized using Raman spectroscopy. As presented in Fig. 1, all of the spectra display two prominent peaks at 1355 and 1587 cm⁻¹, which can be assigned to the D and G bands, respectively. After thermal annealing treatment, the Raman D/G intensity ratios of the Ag–MnO₂/graphene and MnO₂/graphene composites (where the D peak is a defect peak caused by interval scattering,¹⁸ and G refers to the graphene G peak) are lower than that of the as-made graphene, because the thermal reduction actually increased the average size of the crystalline graphene domains in the as-prepared Ag–MnO₂/graphene and MnO₂/graphene composite.¹⁹ Meanwhile, the bands located at 480 and 640 cm⁻¹ in the spectra of the Ag–MnO₂/graphene and MnO₂/graphene composites are contributed to the symmetric stretching vibrations of Mn–O of MnO₂,²⁰ which confirms the presence of the MnO₂ in the as-prepared composite. The peak at the 640 cm⁻¹ in the composite of Ag–MnO₂/graphene increases dramatically compared with MnO₂/graphene and Ag/graphene. This result can be attributed to the surface enhanced Raman scattering from the intense local electromagnetic fields of Ag nanoparticles that accompany Plasmon resonance,²¹ which demonstrates the effective combination of Ag and MnO₂ nanoparticles on graphene sheets.

Fig. 1 Raman spectra of graphene, MnO₂/graphene, Ag/graphene and Ag–MnO₂/graphene composites.

Fig. 2 displays the XRD patterns of graphene, MnO₂/graphene, Ag/graphene and Ag–MnO₂/graphene. For all of the composites, the broadened peaks located at approximately 24.5° are assigned to graphite (002) of the graphene support. Meanwhile, the peaks located at approximately 18.12°, 28.84°, 37.52°, and 60.27° for the MnO₂/graphene, correspond to the (200), (310), (211), and (521) crystalline planes. The main diffraction peaks match well with the standard peaks of α-MnO₂ (JCPDS 44-0141). Meanwhile, the XRD results of Ag–MnO₂/graphene composite show four new diffraction peaks located at about 38.1°, 44.3°, 65.4°, and 77.3°, which can be indexed to the (111), (200), (220), and (311) reflections of the face-centered cubic phase Ag (JCPDS no. 65-2871), the four peaks corresponding well to the peaks of Ag/graphene composite. The results further indicate the successful preparation of Ag and MnO₂ nanoparticles on graphene sheets.

Fig. 2 XRD patterns of graphene, MnO₂/graphene, Ag/graphene and Ag–MnO₂/graphene composites.

TEM images of the as-synthesized Ag–MnO₂/graphene composite are shown in Fig. 3. The nanoparticles cover the surface of graphene sheets with a uniform distribution (Fig. 3a and b). The images show that the nanoparticles with different contrasts can be observed on the graphene sheets. 7.5 wt% of nanoparticles with darker contrasts can be attributed to the Ag particles because silver is a heavy atom and has a strong scattering factor for electrons,²² whereas the nanoparticles with lighter contrasts can be denoted as MnO₂. The high magnification image (the inset in Fig. 3b) shows that the sizes of Ag and MnO₂ nanoparticles are about 6 and 4 nm, respectively. The lattice fringes in the image with a high magnification provide further evidence for distinguishing MnO₂ and Ag nanoparticles. The lattice distance of 0.24 nm that was detected in the darker contrast particles can be assigned to the (111) plane of phase Ag, and the lattice distance of 0.20 nm can be indexed to the (310) plane of MnO₂. The TEM images show that the structural distribution of the nanoparticles offers abundant active sites in electrochemical reactions.

Fig. 3 TEM images (a and b) of Ag–MnO₂/graphene composite and HRTEM (the inset of b) image of Ag and MnO₂ nanoparticles. (c) SEM images of Ag–MnO₂/graphene composite and (d, e and f) the EDX elemental mapping images of Ag–MnO₂/graphene composite.

The content of Ag and MnO₂ can be calculated using EDX to be 7.5 wt% and 32.5 wt%, respectively. The selected overall area of SEM (Fig. 3c) for mapping, as shown in Fig. 3d–f, showed presence of Ag–MnO₂/graphene through distinctive elements, such as C, Mn, and Ag. The elemental mapping of Mn and Ag shows that the amount of silver element has significantly less manganese, which confirms the homogeneous distribution of Ag and MnO₂ nanoparticles. This results is consistent with the TEM results.

The special surface area is 426.4 m² g⁻¹ (Fig. 4) for Ag–MnO₂/graphene, which is larger than that of simplex nano-MnO₂ (35.2 m² g⁻¹).²³ The introduction of Ag nanoparticles and graphene significantly increases the surface area from 35.2 to 426.4 m² g⁻¹. The pore-size distribution (the inset in Fig. 4) is centered at 2–3 nm, which demonstrates that the mesoporous nature of the Ag–MnO₂/graphene composite. The structures provide more chances for oxygen atoms to contact catalyst and enhance the electrolyte contact area, thereby advancing the ORR.

Fig. 4 A typical N₂ adsorption isotherm of the Ag–MnO₂/graphene composite. Inset shows the corresponding pore-size distribution.

XPS was used to probe the chemical states of Mn and Ag in the Ag–MnO₂/graphene composite. The XPS spectrum of Ag–MnO₂/graphene composite (Fig. 5) further confirms the presence of Ag and Mn elements. Fig. 5a shows the XPS spectrum of the Mn 2p signal of Ag–MnO₂/graphene. The Mn 2p XPS spectrum exhibits two major peaks with binding energy values at 653.5 and 642.4 eV, which can be attributed to the Mn 2p_1/2 and Mn 2p_2/3. Whereas the Mn 2p_2/3 peak can be deconvoluted into two peaks at 642.2 and 641.1 eV, which are ascribed to the Mn(IV) and Mn(III) species.²⁴ Fig. 5b presents the typical Ag 3d spectrum of Ag–MnO₂/graphene composite, the peaks of Ag 3d_3/2 and 3d_5/2 can be observed at 374.2 eV and 368.2 eV, respectively, which demonstrates that Ag is present in the metallic state.^25,26 The XPS results agree with that from the Raman and XRD analyses.

Fig. 5 XPS spectra of the Ag–MnO₂/graphene composite: (a) Mn 2p and (b) Ag 3d.

3.2. Electrochemical characteristics of the composites

The CVs for the Ag–MnO₂/graphene composite at 1200 rpm of the rotation rate in a 0.1 M KOH solution has a scanning rate of 5 mV s⁻¹ with N₂-saturated and O₂-saturated, respectively. As shown in Fig. 6a, two existing peaks of the electrode with an Ag–MnO₂/graphene catalyst in O₂-saturated solution for negative scanning are located at approximately 0.16 V and −0.15 V, respectively. However, only a peak of the electrode with Ag–MnO₂/graphene catalyst located approximately 0.16 V in N₂-saturated electrolytes. This result indicates that the oxygen reduction peak of the electrode with Ag–MnO₂/graphene catalyst is observed at −0.15 V. The inset of Fig. 6a shows that a cathodic peak in O₂-saturated electrolyte of the electrode with Pt/C located at approximately −0.19 V is compared with the peak in N₂-saturated electrolyte, which demonstrates that the electrochemical activation of Ag–MnO₂/graphene catalyst to catalyze ORR is more positive than that of Pt/C (−0.19 V). Meanwhile, the current density of the oxygen reduction peak of Ag–MnO₂/graphene catalyst is larger than that of the Pt/C catalyst. The increase in the current density and peak potential for the electrode with Ag–MnO₂/graphene catalyst indicates high catalytic of Ag–MnO₂/graphene toward ORR. An obscure peak at approximately −0.05 V can be attributed to the formation of MnOOH from MnO₂,²⁷ which is masked by the oxygen reduction peak at −0.15 V. Fig. 6b shows the CV of Ag–MnO₂/graphene electrode at different scan rates of 2 mV s⁻¹ to 20 mV s⁻¹. The cathodic peak currents are linear to the square root of the scan rate, as shown the inset in Fig. 6b. These facts indicate a diffusion control for the ORR process, and confirm that the immobilized Ag–MnO₂/graphene is stable.

Fig. 6 (a) CV of the Ag–MnO₂/graphene and Pt/C (the inset of figure a) electrode at scan rate of 5 mV s⁻¹ with the rotation rate at 1200 rpm in a 0.1 M KOH solution saturated with N₂ and O₂, (b) CV of the Ag–MnO₂/graphene electrode at different scan rates with the rotation rate at 1200 rpm in a 0.1 M KOH solution saturated with O₂ and the inset shows the peak current as a function of different rate.

The CV curves of MnO₂/graphene and Ag/graphene composites are studied in a 0.1 M KOH solution with N₂-saturated and O₂-saturated, respectively, to investigated the interaction of silver and manganese dioxide on the catalytic activity of Ag–MnO₂/graphene composite and the catalytic mechanism of oxygen reduction, as shown in Fig. 7a and b. For the CV curve of MnO₂/graphene (Fig. 7a), a pronounced peak at approximately −0.29 V in O₂-saturated solution is contributed to the oxygen reduction of the electrode with MnO₂/graphene composite. Meanwhile, the reduction peak at approximately −0.17 V in N₂-saturated and O₂-saturated solutions is assigned to the IV valence of MnO₂, which is reduced to MnOOH. According to the CV curve of Ag/graphene composite shown in Fig. 7b, the cathodic peak at approximately 0.16 V of the electrode with Ag/graphene composite exists both in O₂-saturated and in N₂-saturated electrolytes, which is assigned to the transformation of silver oxides to metallic silver.⁴ In addition, this condition can also explain the peak at approximately 0.16 V of the electrode with an Ag–MnO₂/graphene composite (Fig. 6a). The reduction peak at approximately −0.35 V of the electrode with Ag/graphene in O₂-saturated is attributed to the oxygen reduction. The catalytic activity of the Ag–MnO₂/graphene catalyst is more positive than that of Ag/graphene and MnO₂/graphene composites. Moreover, the current density of Ag–MnO₂/graphene is larger than that of Ag/graphene and MnO₂/graphene composites because of the synergy of silver and manganese dioxide.

Fig. 7 CVs of the different electrode at scan rate of 5 mV s⁻¹ with the rotation rate at 1200 rpm in a 0.1 M KOH solution saturated with N₂ and O₂: (a) MnO₂/graphene composite, (b) Ag/graphene composite.

According to the above analysis, the catalytic property of Ag–MnO₂/graphene can be attributed to the synergy between silver and manganese dioxide nanoparticles, and the introduction of Ag nanoparticles, which increases the surface area of MnO₂/graphene. The MnO₂/graphene catalyzes the ORR mechanism of the peak at approximately −0.17 V, which could cause reduction of O₂ to HO₂⁻ as described in eqn (1). Thus, the peak at approximately −0.29 V shows the electrochemical transition of HO₂⁻ to OH⁻ (eqn (2)).^28,29

O₂(g) + H₂O(l) + 2e⁻ → HO₂⁻(l) + OH⁻(l) (1)

HO₂⁻(l) + H₂O(l) + 2e⁻ → 3OH⁻(l) (2)

The reaction of eqn (1) can be explained using the mechanism of reaction because manganese dioxide adsorbed the oxygen as shown in the following equation:³⁰

MnOOH(s) + O₂(g) ⇌ MnOOH⋯O₂(s) (3)

MnO₂(s) + H₂O(l) + e⁻⇌ MnOOH(s) + OH⁻(l) (4)

MnOOH⋯O₂(s) + e⁻ ⇌ MnO₂(s) + HO₂⁻(l) (5)

Therefore, the reactions in eqn (3)–(5) is equal to the eqn (1), which corresponds to the common knowledge that O₂ reduction occurs simultaneously with MnO₂ reduction.

Therefore, the oxygen reduction peak of the electrode with an Ag–MnO₂/graphene catalyst is because the insertion of a proton into MnO₂ as the eqn (3) and (4), the eqn (5) no longer appears in this process, which then produces synergistic effect with Ag, and the MnOOH adsorption of O₂ transfer to the surface of Ag, which can be explained using the mechanism of reaction, as shown in the following equations:

MnOOH⋯O₂(s) + Ag(s) ⇌ MnOOH(s) + Ag⋯O₂(s) (6)

2Ag⋯O₂(s–g) + H₂O(l) + 2e⁻ → Ag₂O(s) + 2HO₂⁻(l) (7)

Ag may further facilitate the electrochemical reduction of HO₂⁻ as eqn (2), thus the total reaction mechanism of the electrode with Ag–MnO₂/graphene at −0.05 V is:

2Ag(s) + 2MnO₂(s) + 3H₂O(l) + 2O₂(g) + 4e⁻ → Ag₂O(s) + 2MnOOH(s) + 2OH⁻(l) + 2HO₂⁻(l) (8)

The instability of Ag₂O and MnOOH, allows these elements easily decompose into Ag and MnO₂, which continue catalyze the ORR circularly. Thus, the catalyst of Ag–MnO₂/graphene catalyzes O₂ to OH⁻ at −0.15 V. This result can be attributed to eqn (8), which is equal to the combination of eqn (1) and (2).

The LSV of the Ag–MnO₂/graphene composite was investigated between −0.7 V and 0.1 Vat a scan rate of 1 mV s⁻¹ in O₂-saturated 0.1 M KOH solution with a rotation rate of 2400 rpm, as shown in Fig. 8a. The ORR limiting current density for the Ag–MnO₂/graphene catalyst is approximately 5.62 mA cm⁻², which is larger than that of the Ag/graphene composite (0.92 mA cm⁻²), and MnO₂/graphene composite (2.82 mA cm⁻²), but similar to that of the Pt/C (5.83 mA cm⁻²). The results indicate that Ag effectively improves electrical conductivity. Another critical factor determining the performance of a catalyst is the onset potential. The onset potential of Ag–MnO₂/graphene was detected at 0.068 V, which is close to that of Pt/C (0.079 V). The Tafel slope of Ag–MnO₂/graphene and Pt/C catalyst is 86 and 74 mV per decade, respectively, as shown in Fig. 8b. The low ORR Tafel slope of Ag–MnO₂/graphene implies a transition from the Langmuirian absorption to the Temkin adsorption of adsorbed O/OH groups.³¹ Therefore, the electrochemical catalyst activation of Ag–MnO₂/graphene composite is close to that of Pt/C for ORR.

Fig. 8 LSV curves (a) of the Ag/graphene, MnO₂/graphene, Ag–MnO₂/graphene and Pt/C catalysts in a 0.1 M KOH solution with O₂-saturated at 2400 rpm with the scanning rate of 1 mV s⁻¹, (b) the corresponding Tafel plots. (c) LSV curves of the Ag–MnO₂/graphene composite in a 0.1 M KOH solution with O₂-saturated at different rotation rate with the scanning rate of 1 mV s⁻¹ and the inset shows the corresponding Koutecky–Levich plots at potentials of −0.4, −0.5, and 0.6 V, respectively.

The LSV of Ag–MnO₂/graphene at different rotation rates from 400–2400 rpm are shown in Fig. 8c. The RDE data which were obtained through the Koutecky–Lecich equation are as follows.^31,32

i_k = nFkC₀

i_d = 0.62nFD₀^2/3v^−1/6C₀ω^1/2

where i_k and i_d represent the kinetics and diffusion limiting current densities (Am⁻²), respectively. n is the electron number involved in ORR; F is the Faraday constant (96 485 C mol⁻¹); k is the rate constant for ORR (m s⁻¹); D₀ is the diffusion coefficient of O₂ in solution (1.86 × 10⁻⁵ cm² s⁻¹ in 0.1 M KOH); C₀ is the saturation concentration of O₂ in solution (1.2 mM m⁻³ in 0.1 M KOH); ν is the kinematic viscosity (0.0100810⁻⁶ cm² s⁻¹ in 0.1 M KOH); and ω is the angular frequency of the rotation (rad s⁻¹). The electron transfer number (n) during ORR was calculated to be 3.5, 3.7, and 3.9 at −0.4 V, −0.5 V, and −0.6 V, respectively. The reaction mechanism of the near-four electron transfer pathway for the Ag–MnO₂/graphene catalyst further proves eqn (8).^29,30 The incorporation of Ag in α-MnO₂ improves the process of ORR efficiently.

In Fig. 9, the impedance spectra of the MnO₂/graphene, Ag/graphene and Ag–MnO₂/graphene electrodes are consisted of one semicircle and a slope. Inset shows the equivalent circuit of the three electrodes, where R₀, R₁, R₂, CPE1, CPE2 and CPE3 are denoted as solution resistance, adsorption resistance, electrochemical reaction resistance, adsorption capacitance, double layer capacitance and diffusion capacitance, respectively. Because of the porous morphology of electrodes, a constant phase element (CPE) replaces the capacitor element reasonably. A CPE is usually defined as {Y(jw)}⁻¹. A nonlinear, least-square fitting calculation is performed using the equivalent circuit. For Ag/graphene electrode, the R₂ and CPE2 value (T) are 0.618 Ω g and 1.812 F g⁻¹, MnO₂/graphene electrode are 0.201 Ω g and 2.061 F g⁻¹, and Ag–MnO₂/graphene electrode 0.107 Ω g and 4.784 F g⁻¹. The R₂ of the MnO₂/graphene electrode is lower than that of the MnO₂/graphene and Ag/graphene electrodes. This further confirms that the electrical conductivity of Ag–MnO₂/graphene composite is improved significantly by the introduction of Ag nanoparticles, and the results consistent well with the polarization curves and CVs. The value of CPE2 is related to the true reaction areas of the electrode, so the double layer capacitance of Ag–MnO₂/graphene electrode is larger than that of the MnO₂/graphene and Ag/graphene electrodes, which means Ag–MnO₂/graphene electrode has larger reaction area.

Fig. 9 The EIS results of Ag/graphene, MnO₂/graphene and Ag–MnO₂/graphene in 0.1 M KOH solution and the equivalent circuit of the electrodes (inset) 298 K.

4. Conclusion

The Ag–MnO₂/graphene composite with a mesoporous structure was successfully prepared using immersion–calcination method. The Ag and MnO₂ nanoparticles are uniformly distribution on the graphene sheets. The CV curves show that the oxygen reduction potential of the Ag–MnO₂/graphene catalyst shifts positively by 200 and 140 mV compared with that of Ag/graphene and MnO₂/graphene composites, respectively, and even shift positively (40 mV) compared with that of 20% Pt/C. The activity of the Ag–MnO₂/graphene composite improved compared with that of 20% Pt/C. The results imply that the Ag–MnO₂/graphene composite is a promising catalyst for fuel cells in alkaline solution.

Acknowledgements

The authors gratefully acknowledge the financial support from the Open Project of Key Lab Adv. Energy Mat. Chem. (Nankai Univ.) (KLAEMC-OP201201) partially.

References

1. F. Y. Cheng, Chem. Soc. Rev., 2012, 41, 2172–2192.
2. V. Neburchilov, H. Wang, J. J. Martin and W. Qu, J. Power Sources, 2010, 195, 1271–1291.
3. R. Chen, H. Li, D. Chu and G. Wang, J. Phys. Chem. C, 2009, 113, 20689–20697.
4. J. Guo, A. Hsu, D. Chu and R. Chen, J. Phys. Chem. C, 2010, 114, 4324–4330.
5. M. V. Ananth, V. V. Giridhar and K. Renuga, Int. J. Hydrogen Energy, 2009, 34, 658–664.
6. T. Burchardt, P. Gouérec, E. Sanchez-Cortezon, Z. Karichev and J. H. Miners, Fuel, 2002, 81, 2151–2155.
7. Y. G. Sun and Y. N. Xia, Science, 2002, 298, 2176–2179.
8. L. Z. Yuan, L. H. Jiang, J. Liu, Z. X. Xia, S. L. Wang and G. Q. Sun, Electrochim. Acta, 2014, 135, 168–174.
9. F. Cheng, Y. Su, J. Liang, Z. Tao and J. Chen, Chem. Mater., 2009, 22, 898–905.
10. R. Zou, Z. Zhang, L. Yu, Q. Tian, Z. Chen and J. Hu, Chem.–Eur. J., 2011, 17, 13912–13917.
11. J. Zhu and J. He, ACS Appl. Mater. Interfaces, 2012, 4, 1770–1776.
12. F. Y. Cheng, T. R. Zhang, J. Du, X. P. Han, C. C. Chen and J. Chen, Angew. Chem., Int. Ed., 2013, 52, 2474–2477.
13. Y. L. Cao, H. X. Yang, X. P. Ai and L. F. J. Xiao, Electroanal. Chem., 2003, 557, 127–134.
14. M. A. Kostowskyj, D. W. Kirk and S. J. Thorpe, Int. J. Hydrogen Energy, 2010, 35, 5666–5672.
15. F. P. Hu, X. G. Zhang, F. Xiao and J. L. Zhang, Carbon, 2005, 43, 2931–2936.
16. D. H. Wang, R. Kou, D. W. Choi, Z. G. Ynag, Z. M. Nie, J. Li, L. V. Saraf, D. H. Hu, J. Zhang, G. L. Graff, J. Liu, M. A. Pope and I. A. Aksay, ACS Nano, 2010, 4, 1587–1595.
17. C. D. Zangmeister, Chem. Mater., 2010, 22, 5625–5629.
18. A. C. Ferrari, Solid State Commun., 2007, 143, 47–57.
19. H. L. Wang, J. T. Robinson, X. L. Li and H. J. Dai, J. Am. Chem. Soc., 2009, 131, 9910–9911.
20. T. Gao, M. Glerup, F. Krumeich, R. Nesper, H. Fjellvag and P. Norby, J. Phys. Chem. C, 2008, 112, 13134–13140.
21. R. Pasricha, A. Gupta and A. K. Srivastava, Small, 2009, 5, 2253–2259.
22. T. Akita, M. Kohyama and M. Haruta, Acc. Chem. Res., 2013, 46, 1773–1782.
23. S. Y. Wang, J. N. Xie, T. R. Zhang and V. K. Varadan, J. Power Sources, 2009, 186, 523–538.
24. A. S. Reddy, C. S. Gopinath and S. Chilukuri, J. Catal., 2006, 243, 278–291.
25. Y. Z. Zhou, J. Yang, T. T. He, H. F. Shi, X. N. Chen and Y. X. Lu, Small, 2013, 9, 3445–3454.
26. A. M. Ferraria, S. Boufi, N. Battaglini, A. M. B. Rego and M. Reivilar, Langmuir, 2010, 26, 1996–2001.
27. F. H. B. Lima, M. L. Calegaro and E. A. Ticianeli, Electrochim. Acta, 2007, 52, 3732–3738.
28. A. Verma, A. K. Jha and S. Basu, J. Power Sources, 2005, 141, 30–34.
29. F. Y. Cheng, Y. Su, J. Liang and Z. L. Tao, Chem. Mater., 2010, 22, 898–905.
30. H. L. Gao, Z. Y. Li and Q. Xue, J. Power Sources, 2014, 248, 565–569.
31. F. W. T. Goha, Z. L. Liu, X. M. Ge, Y. Zong, G. J. Du and T. S. A. Hora, Electrochim. Acta, 2013, 114, 598–604.
32. F. P. Hu, X. G. Zhang, F. Xiao and J. L. Zhang, Carbon, 2005, 43, 2931–2936.

---