Study of carbon black supported amorphous Ni–B nano-catalyst for hydrazine electrooxidation in alkaline media

Abstract

Carbon black supported amorphous Ni–B alloy catalyst (Ni–B/C) was prepared by a simple chemical reduction method and applied as an anode catalyst for direct hydrazine fuel cells (DHFCs). The structure and morphology of the obtained Ni–B/C nano-particles (NPs) were characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD), inductively coupled plasma optical emission spectrometry (ICP-OES) and X-ray photoelectron spectroscopy (XPS). The results showed that Ni–B/C was in an amorphous state and the NPs were highly dispersed on the surface of carbon black. The electrocatalytic performance of amorphous Ni–B/C towards hydrazine electrooxidation was studied by cyclic voltammetry (CV), linear sweep voltammetry (LSV) and chronoamperometry (CA) methods using a rotating disk electrode (RDE) and compared with crystalline Ni/C NPs, where a two-fold increase in the catalytic current density towards hydrazine electrooxidation was observed.

---

Introduction

Direct hydrazine fuel cells (DHFCs) have been widely investigated because of their advantages, which include zero carbon emissions (i.e. only N₂ and H₂O production), a high theoretical potential (1.64 V), high theoretical energy density, easy storage, and in particular, the possibility of using non-noble metals as anode and cathode catalysts.¹ Different transition metal alloys such as Co and Ni alloys were found to be active for hydrazine electrooxidation in alkaline media by combinatorial exploration² and studies regarding the catalytic performance of transition metal and transition metal alloys for hydrazine electrooxidation have drawn significant attention. Using metal disk electrodes, Asazawa et al.³ examined the electrooxidation of hydrazine on different metals and observed that Ni and Co were highly active toward the electrooxidation of hydrazine. Sakamoto et al.⁴ reported a high-performance Ni–La catalyst for the electrooxidation of hydrazine hydrate in alkaline media. An aerosol-derived Ni_1−xZn_x electrocatalyst was synthesized and its hydrazine electrooxidation activity was studied by Martinez and co-workers,⁵ where a high activity for hydrazine electrooxidation was observed. Sakamoto and colleagues⁶ synthesized Ni–Co alloy nano-catalysts using a liquid reduction process and evaluated the electrocatalytic activities using a rotating disk electrode (RDE) method. They found Ni_0.50Co_0.50 exhibited higher activity than pure Ni, Co catalysts for the anodic oxidation of hydrazine in alkaline solution. The binary and ternary crystalline Ni alloys have already been proved to be very active for hydrazine electrooxidation in alkaline media by recent researches;^2–6 however, the study of amorphous Ni alloys for hydrazine electrooxidation in alkaline media was very limited.

In recent years, ultrafine amorphous alloys have attracted substantial attention. Combining the features of amorphous structure and ultrafine size, they exhibited large surface areas and a high concentration of highly coordinated unsaturated sites.^7–9 Previous studies expected the special properties of amorphous alloys could lead to better catalytic activities. In fact, their high catalytic activities have already been proved in a lot of fields such as hydrogenation,^10–12 dehydrogenation reactions;¹³ and some alcohol electrooxidation reactions,^14,15 but still, the study of their performance for hydrazine electrooxidation in alkaline media was very rare. In the 1970s, nickel boride was proposed as an anode catalyst in DHFCs;^16,17 however, the literature only contained the performance of Ni–B hydrazine fuel cells with Ni–B synthesized in situ on the anodes, and no detailed studies concerning the performance of nano-scale amorphous Ni–B/C as catalysts for the electrooxidation of hydrazine was reported.

Here amorphous Ni–B/C NPs were synthesized using a chemical reduction method and their catalytic activities towards hydrazine electrooxidation were investigated. The current density generated from hydrazine electrooxidation by amorphous Ni–B/C was approximately twice as high as that generated by Ni/C in our study.

Experimental

Catalyst preparation

Carbon black supported amorphous Ni–B alloy NPs were prepared through a chemical reduction method.¹⁸ 0.30 g carbon black (Vulcan® XC-72, Cabot, American) and 0.81 g NiCl₂·6H₂O (Beijing Chemical Factory) were fully mixed in 50.0 mL ethylene glycol (Beijing Chemical Factory) and an approximate loading of 40% was achieved. After being sonicated for 20 min, the mixture was put in an ice-water bath. Then 50.0 mL NaBH₄ (Beijing Chemical Factory) solution (0.01 g mL⁻¹) was added to the nickel salt–carbon black mixture via a constant flow pump with a flow rate of 5 mL min⁻¹. During the whole process, the solution was put in ice-water bath under constant mechanical stirring. After being rinsed by deionized water 3–5 times, the prepared catalysts were stored in deionized water before further characterization.

Carbon black supported nickel particles were also prepared to compare the catalytic activity with Ni–B/C NPs. 0.30 g carbon black and 0.81 g NiCl₂·6H₂O salts were mixed in 50.0 mL ethylene glycol. After being sonicated for 20 min, the mixture was kept in a water bath of 70 °C with constant mechanical stirring. Then 10.0 mL hydrazine (85 wt%, Beijing Chemical Factory) was added into the mixture with a constant flow pump at a flow rate of 1.0 mL min⁻¹. After then, 10.0 mL 10 M KOH solution was added into the mixture, and the Ni/C NPs were obtained after stirring the suspension for another 15 min. After being rinsed by deionized water 3–5 times, the Ni/C NPs were also stored in water for further characterization. All catalysts were dried in a vacuum drier and ground before characterization.

Physical characterization

The XRD characterization was conducted with an X-ray diffractometer (SHIMADZU, XRD-6000). The Cu Kα excitation source was operated at a potential of 40 kV and a current of 30 mA. 2θ diffraction angles in this research ranged from 20° to 85°.

The microstructure of as prepared catalyst was further studied by transmission electron microscopy (TEM), selected area electron diffraction (SAED) with a transmission electron microscope (JEOL, JEM-2100F). The accelerating voltage was 200 kV. The surface of prepared Ni–B/C and Ni/C was characterized by energy dispersive X-ray spectroscopy (EDS) with an energy dispersive spectrometer (EDAX, GENESIS).

The existence of Ni and B was confirmed with an inductively coupled plasma optical emission spectrometer (Perkin Elmer, OPTIMA 2000). In brief, 100.0 mg catalyst was digested with a mixed acid of hydrochloric acid and nitric acid. After a standard line was obtained, the samples were injected into the instrument and the concentration of Ni and B were calculated according to the results obtained.

To understand the elements and their existing states on the surface of prepared Ni–B/C and Ni/C catalysts, the XPS spectra were taken with an X-ray photoelectron spectroscope (Thermo Scientific, ESCALAB 250Xi). The excitation source was the Kα radiation of an aluminum anode. The pass energy was set to 30 eV, the step size was 50 meV.

Electrochemical characterization

Electrochemical tests were conducted using a potentiostat (Shanghai Chenhua Co. Ltd, CHI 660c) and a rotating disk electrode (RDE, Pine Instrument, AFMSRCE). Cyclic voltammetry (CV) and chronoamperometry (CA) tests were conducted with a three-electrode system. The working electrode was a glass carbon electrode (GCE, diameter = 5 mm) modified with prepared catalyst inks. The counter electrode was a self-prepared Pt wire electrode. The reference electrode was a self-prepared reversible hydrogen electrode (RHE) at pH 14. The rotation speed of RDE was set to 900 rpm to facilitate the diffusion of hydrazine.

Catalyst inks were prepared as following: 5.0 mg catalyst was added into 1.0 mL 0.5 wt% Nafion–isopropyl alcohol solution. The suspension was sonicated to disperse the catalyst powder in solution and then catalyst ink was acquired. 2.5 µL of the catalyst ink was then added onto the GCE and left to dry under an infrared lamp.

1 M KOH solution was used as the electrolyte. The electrolytes were bubbled with nitrogen for 30 min to remove dissolved oxygen and maintained under nitrogen atmosphere during all the electrochemical tests. Before all the electrochemical tests, the prepared catalyst on the GCE was activated by repeating CV scanning from −0.2 V to 0 V (vs. RHE) with a scan rate of 50 mV s⁻¹. All measurements were conducted under room temperature unless specifically denoted.

Results and discussion

Physical characterization

The XRD patterns of the prepared amorphous Ni–B/C and crystalline Ni/C were presented in Fig. 1. The pattern of Ni–B/C exhibited a broad peak at 2θ = 45°, which is a typical feature of amorphous nickel alloys.^13,19 The XRD pattern of Ni/C showed typical face centered cubic (fcc) peaks. These diffraction peaks in Fig. 1 at approximately 2θ = 45°, 52° and 76° were indexed to the (111), (200) and (220) crystalline plane diffraction of Ni, respectively. The size of prepared crystalline Ni supported on carbon black was calculated according to the Scherrer formula, which was approximately 10 nm.

Fig. 1 XRD patterns of the amorphous Ni–B/C and crystalline Ni/C.

Fig. 2a presented a TEM image of the prepared Ni–B/C, where the Ni–B particles supported on carbon black were indicated. It depicted the Ni–B particles were well-dispersed on carbon black and smaller than the Ni particles, as shown in Fig. 2b. The selected area electron diffraction (SAED) pattern corresponding to the observed Ni–B/C particles was presented in the inset of Fig. 2a. There were only concentric circles observed, confirming the amorphous state of the prepared Ni–B/C. As comparison, the diffraction spots were observed in the SAED images of Ni/C as presented in the inset of Fig. 2b, demonstrating that Ni/C was in a crystalline state. The size and morphology of carbon black in Fig. 2a were identical to the TEM image of bare carbon black (not shown) and generated no signal of Ni in the EDS data, while the signal of Ni in the EDS data of Ni–B/C was very strong, as presented in Fig. 2c. Furthermore, it can be also observed from Fig. 2a the Ni–B particles grown into separate small particles and were much better dispersed on carbon black than Ni, suggesting that the alloying of B may facilitate the dispersion of Ni–B. The average particle size of Ni–B/C was estimated to be 7 nm according to the TEM image and the particle size distribution was presented in Fig. 2d, which indicated that the particle size of Ni–B/C was much smaller than that of Ni/C around 10 nm according to the XRD results as shown in Fig. 1.

Fig. 2 TEM and SAED characterization of prepared Ni/C and Ni–B/C catalysts. (a) TEM image of Ni–B/C; inset of (a): SAED image of Ni–B/C. (b) TEM image of Ni/C; inset of (b): SAED image of Ni/C. (c) EDS data of Ni–B/C. (d) Particle size distribution of Ni–B/C.

The ICP-OES study was performed to confirm the successful preparation of Ni–B and acquire the bulk composition of prepared Ni–B/C catalyst. The results indicated that 100.0 mg of Ni–B/C contained approximately 33.0 mg of Ni and 4.4 mg of B, whereas 100.0 mg of Ni/C contained approximately 38.7 mg of Ni and negligible B, which also indicated the successful alloying of B in Ni–B/C.

The XPS characterization was performed to further prove that B was successfully alloyed into Ni in the prepared Ni–B/C and get a deeper understanding of the surface states of prepared Ni–B/C and Ni/C catalysts. The results were shown in Fig. 3. The full XPS spectra of prepared Ni–B/C and Ni/C catalysts were presented in Fig. 3a. It can be seen that the spectra of Ni–B/C and Ni/C catalysts were very similar except that a small signal appeared in the XPS spectrum of Ni–B/C in the B binding energy (B.E.) region (188–194 eV), whereas no such signal was observed in the spectrum of Ni/C. The B binding energy regions of both Ni–B/C and Ni/C catalysts were magnified and shown in Fig. 3b. The deconvolution of the B signals for Ni–B/C gave two peaks with binding energies of 188.4 eV and 192.0 eV, which could be assigned to elemental B(0) and oxidized B(III), respectively. Also, it can be observed that the B 1s binding energy of elemental B had moved 1.3 eV towards the high energy side compared to pure B, which indicated that Ni–B was successfully prepared and the electrons transferred from B to Ni in synthesized Ni–B catalyst.²⁰ The surface states of Ni in Ni–B/C and Ni/C were also analyzed from the XPS data, as shown in Fig 3c. The peaks at approximately 853.2 eV were assigned to the 2p_3/2 orbitals of Ni–B and Ni, and the peaks at 855.0 eV and 862.5 eV were assigned to the 2p_3/2 orbital of Ni(OH)₂ and satellite peaks of the Ni(OH)₂ 2p_3/2 orbital, respectively. The appearance of oxidized B and oxidized Ni species are the results of the air exposure before the ex situ XPS characterization. So in this study, in order to avoid air exposure of Ni–B and Ni catalyst and prevent them from being oxidized, the prepared Ni–B/C and Ni/C catalysts were stored in water and dried in a vacuum drier to avoid oxidation before the electrochemical tests. In addition, by comparing the standard reduction potentials of N₂/N₂H₄ and Ni(OH)₂/Ni_(s), it is apparent that the Ni(II) will be reduced to Ni(0) during the electrochemical tests even if the Ni–B/C and Ni/B were oxidized by oxygen. As a result, the reduced Ni(0) in both Ni–B/C and Ni/C served as active species in the hydrazine electrooxidation process. Therefore the Ni(0) were considered to be responsible for hydrazine electrooxidation in Ni–B/C catalyst.

Fig. 3 XPS spectrum of prepared Ni–B/C and Ni/B catalysts. (a) Full XPS spectrum of Ni–B/C and Ni/B. (b) B 1s orbital binding energy of Ni–B/C and Ni/B. (c) Ni 2p_3/2 orbital binding energy region of Ni–B/C and Ni/B.

Electrochemical tests

CV scanning was performed in 1 M KOH solution for both Ni–B/C and Ni/C to investigate their electrochemical behavior. The scanning was performed from −0.2 V to 0.6 V (vs. RHE) at the scanning speed of 20 mV s⁻¹. The CV curves were presented in Fig. 4a. It can be seen from the CV curves that the oxidation peak and the reduction peak corresponding to Ni oxidation–reduction reaction emerged, which indicated that the Ni–B/C and Ni/C catalysts were oxidized when the potential exceeded 0.3 V (vs. RHE). As a result, the potential should be kept below 0.2 V (vs. RHE) to avoid oxidation of the catalysts when testing the electrocatalytic activity of Ni–B/C and Ni/C catalysts. From the CV curves it was also observed that the redox current density generated by Ni–B/C was much larger than Ni/C catalysts, suggesting a larger electrochemical activity surface area of Ni–B/C compared with that of Ni/C. It was also noticed that the reduction peak moved negatively in Ni–B/C compared with Ni/C catalyst, suggesting that it was more difficult for Ni–B/C to accept electrons, which can be explained by the transportation of electrons from B to Ni through B alloying in Ni–B/C. This was supported by the XPS data discussed later. Fig. 4b presents the CV curves of Ni–B/C and Ni/C in 1 M KOH solution containing 1 M hydrazine hydrate. It can be observed that both Ni–B/C and Ni/C exhibited electrochemical activities towards hydrazine electrooxidation, and hydrazine could be electrooxidized by both Ni–B/C and Ni/C catalysts when the potential exceeded −0.2 V (vs. RHE).

Fig. 4 Electrochemical tests of prepared Ni–B/C and Ni/C catalysts. (a) CV curves of Ni–B/C and Ni/C in bare 1 M KOH solution. (b) CV curves of Ni–B/C and Ni/C in 1 M KOH and 1 M hydrazine. Scan rate: 20 mV s⁻¹. RDE: 900 rpm.

To compare the electrocatalytic activity of Ni/C and Ni–B/C towards hydrazine electrooxidation, the linear sweep voltammetry (LSV) curves were obtained in 1 M KOH containing 20 mM hydrazine hydrate using a RDE at a scan rate of 20 mV s⁻¹ and a rotation speed of 900 rpm, which were presented in Fig. 5a. The results were normalized to the mass of Ni in Ni–B/C and Ni/C because Ni was generally considered as the active site.¹⁰ As evident in Fig. 5a, the mass current density of the prepared Ni–B/C reached 1495 A g⁻¹ Ni under the potential of 0.2 V (vs. RHE), which was approximately twice higher than that of Ni/C (500 A g⁻¹ Ni at 0.2 V vs. RHE). Also, the onset potential for hydrazine electrooxidation of Ni–B/C was about 0.05 V lower than that of Ni/C. In addition, the current density of hydrazine electrooxidation under 60 °C (see Fig. 6a) was obtained and it was much higher than that of several Ni alloy catalysts in literature, which were compared and listed in Table 1. These results demonstrate that Ni–B/C was more active than the aforementioned catalysts. Although the catalytic activity of Ni–B/C was still inferior to that of Pt and Pd catalysts, its low cost provides access to new anode catalysts for DHFCs.

Fig. 5 (a): LSV curves of Ni–B/C and Ni/C in 1 M KOH containing 20 mM hydrazine. (b) CA curves of Ni–B/C and Ni/C in 1 M KOH with 20 mM hydrazine. Potential: 0.2 V (vs. RHE).

Fig. 6 (a) and (b) LSV curves of hydrazine electrooxidation on Ni–B/C and Ni/C in 1.0 M KOH with 20.0 mM N₂H₄ under different temperatures. (c) Plot of the logarithms of the current densities at 0.2 V (vs. RHE) against the absolute temperatures (1/T).

Table 1 Comparison of electrocatalytic performances of different catalysts

Catalyst	Hydrazine concentration	Current density (A g^−1 total metal)	Potential (V vs. RHE)	Ref.
Ni0.5Co0.5	100 mM	1300	0.22	3
Ni0.87Zn0.13	5%	440	0.05	5
Ni0.9La0.1	1 M	759.5	0.221	4
Ni–B	20 mM	2780	0.20	This work

---

The electrocatalytic performance of Ni–B/C and Ni/C was also compared by chronoamperometry (CA) tests. The tests were also conducted in 1 M KOH solution containing 20 mM hydrazine as presented in Fig. 5b. The much higher current density generated by Ni–B/C demonstrated that Ni–B/C was more active than Ni/C for hydrazine electrooxidation.

The effect of temperature on the electrooxidation of hydrazine was studied to explore the kinetic property of the Ni–B/C and Ni/C catalyzed hydrazine electrooxidation reactions. The tests were carried out in 1 M KOH containing 20 mM hydrazine hydrate, which was heated in a water bath at different temperatures. The LSV curves for Ni–B/C and Ni/C catalysts were obtained at 40, 50, 60, 70 and 80 °C, as shown in Fig. 6a and b, respectively. The current density of hydrazine electrooxidation became higher with increasing temperature, indicating that the hydrazine oxidation process was an endothermic reaction. The logarithms of the current densities at 0.2 V (vs. RHE) were plotted as a function of the reciprocal of the absolute temperatures (1/T), which exhibited good linear relationships as shown in Fig. 6c. According to the Arrhenius equation, the activation energy of hydrazine electrooxidation at 0.2 V (vs. RHE) were calculated to be 13.83 kJ mol⁻¹ and 20.67 kJ mol⁻¹ (at 0.2 V vs. RHE) for Ni–B/C and Ni/C, respectively. The activation energy of hydrazine electrooxidation on Ni–B/C was 33% lower than that on Ni/C, which indicated that the activity of Ni–B/C NPs were much higher than that of Ni/C NPs.

The durability of Ni–B/C was tested by accelerated ageing tests. During an accelerated ageing test, the prepared Ni–B/C catalyst was repeatedly scanned for 1000 cycles in 1 M KOH solution. The LSV curves of the prepared Ni–B/C catalyst in 1 M KOH containing 20 mM hydrazine before and after cycling was compared to see the effect to the performance of hydrazine electrooxidation. It can be concluded from Fig. 7 that the amorphous Ni–B/C exhibited high stability during the test because the LSV curves of hydrazine electrooxidation catalyzed by Ni–B/C before and after the accelerated ageing test were almost identical.

Fig. 7 LSV curves of Ni–B/C in 1 M KOH containing 20 mM hydrazine before and after accelerated aging test.

From the results of electrochemical tests performed, it can be concluded that the prepared Ni–B/C showed a much higher electrocatalytic activity towards the electrooxidation of hydrazine in alkaline media than the Ni/C catalyst and two reasons were accounted for this phenomenon. The first one was that the B alloying can facilitate the dispersion of Ni supported on carbon black, which was supported by the XRD and TEM data presented in the physical characterization section. According to the XRD data, the diameter of prepared Ni on carbon black was about 10 nm, while the TEM data showed that the Ni–B particles had an average diameter of 7 nm. The TEM image of Ni/C further confirmed that B alloying made the Ni–B particles better dispersed than Ni particles on carbon black, while Ni particles agglomerated severely. Another reason was that the alloying of B can change the electronic states of Ni, which may improve the catalytic activity of Ni atoms in prepared Ni–B/C catalyst, and this was explained with the help of XPS data. It was observed from the XPS spectrum that the binding energy of B 1s orbital of elemental B had moved from 187.1 eV to 188.4 eV in Ni–B/C. The increase of binding energy of B(0) indicated that the electrons in Ni–B/C were transferred from B atoms to Ni atoms in Ni–B. With the Fermi level of Ni more enriched with electrons, the catalytic performance of Ni will be improved. Since Ni(0) were considered as the active sites in Ni/C and Ni–B/C catalysts as discussed above, the change of electronic states of Ni in Ni–B/C caused by B alloying was also responsible for the better electrocatalytic activity of Ni–B/C.

Conclusions

A carbon-supported amorphous Ni–B alloy catalyst for hydrazine electrooxidation was successfully prepared using a facile method. The TEM and XRD data revealed that the Ni–B particles were homogeneously dispersed on the carbon black, and the XPS results showed that B alloying can affect the electronic states of Ni in Ni–B/C catalysts and improve the catalytic activity of Ni in Ni–B/C catalyst. The Ni–B/C catalyst exhibited a much higher current density than Ni/C and many other reported Ni alloy catalysts towards hydrazine electrooxidation in alkaline media and also showed high stability during the electrochemical tests, which makes amorphous Ni–B/C catalyst a very promising anode catalyst for DHFCs.

Acknowledgements

This work was financially supported by grants from the National Natural Science Foundation of China (nos 21003007, U1137602), the National High Technology Research and Development Program of China (2013AA031902), National Program on Key Basic Research Project (no. 2011CB935700), the National Science Foundation of Beijing (no. 2132051), Beijing Higher Education Young Elite Teacher Project (no. 29201493) and the Fundamental Research Funds for the Central Universities (YWF-13-T-RSC-030).

References

1. A. Serov and C. Kwak, Appl. Catal., B, 2010, 98, 1–9.
2. T. Sakamoto, K. Asazawa, K. Yamada and H. Tanaka, Catal. Today, 2011, 164, 181–185.
3. K. Asazawa, K. Yamada, H. Tanaka, M. Taniguchi and K. Oguro, J. Power Sources, 2009, 191, 362–365.
4. T. Sakamoto, K. Asazawa, U. Martinez, B. Halevi, T. Suzuki, S. Arai, D. Matsumura, Y. Nishihata, P. Atanassov and H. Tanaka, J. Power Sources, 2013, 234, 252–259.
5. U. Martinez, K. Asazawa, B. Halevi, A. Falase, B. Kiefer, A. Serov, M. Padilla, T. Olson, A. Datye, H. Tanaka and P. Atanassov, Phys. Chem. Chem. Phys., 2012, 14, 5512–5517.
6. J. Sanabria-Chinchilla, K. Asazawa, T. Sakamoto, K. Yamada, H. Tanaka and P. Strasser, J. Am. Chem. Soc., 2011, 133, 5425–5431.
7. J. V. Wonterghem, S. Morup, J. W. Christion, S. Charles and W. S. Wells, Nature, 1986, 322, 622–623.
8. C. Wu, Y. Bai, X. Wang, F. Wu and C.-Z. Zhang, Solid State Ionics, 2008, 179, 924–927.
9. Y. G. He, M. H. Qiao, H. R. Hu, Y. Pei, H. X. Li, J. F. Deng and K. N. Fan, Mater. Lett., 2002, 56, 952–957.
10. H. X. Li, H. Li, W. L. Dai and M. H. Qiao, Appl. Catal., A, 2003, 238, 119–130.
11. S.-W. Yan, H. Fan, C. Liang and Z. Li, Chem. Res. Chin. Univ., 2012, 33, 2067–2073.
12. D. Xu, H. Wang, Q. Guo and S. Ji, Fuel Process. Technol., 2011, 92, 1606–1610.
13. D. Xue, H. Chen, G. H. Wu and J. F. Deng, Appl. Catal., A, 2001, 214, 87–94.
14. H. Cao, Z. Wang, G. Hou and G. Zheng, Surf. Coat. Technol., 2010, 205, 885–889.
15. Y. Cai, Y. Hou and J. Guo, in Progress in Renewable and Sustainable Energy, Pts 1 and 2, ed. Y. G. Li, Y. Li and W. G. Pan, 2013, pp. 929–933.
16. S. Meibuhr and E. Zeitner, US Pat., 3852116 A, 1974.
17. M. Salomon, US Pat., 3513028 A, 1970.
18. Y. Chen, Catal. Today, 1998, 44, 3–16.
19. G. Bai, L. Niu, M. Qiu, F. He, X. Fan, H. Dou and X. Zhang, Catal. Commun., 2010, 12, 212–216.
20. J. Qiao, L. Xu, Y. Liu, P. Xu, J. Shi, S. Liu and B. Tian, Electrochim. Acta, 2013, 96, 298–305.

---