A highly active hydrazine fuel cell catalyst consisting of a Ni–Fe nanoparticle alloy plated on carbon materials by pulse reversal

Abstract

A noble metal-free, highly active electrocatalyst for hydrazine fuel cells has been prepared by pulse reversal plating. Plating Ni–Fe alloy on carbon materials (multi-walled carbon nanotubes and graphene) to produce electrocatalysts is facile, rapid, proceeds at low temperature, and does not damage the sp² hybridization of the carbon support.

---

As a potential reactor in fuel cells, hydrazine (N₂H₄) has unique advantages including superior theoretical standard equilibrium potential, high power density and environmentally benign products,¹ all of which sparked our interest in developing high-power density direct hydrazine–air fuel cells. Recent efforts to improve performance of catalysts in fuel cells have focused on using noble metal-free catalysts,^1,2 adjusting their composition,¹ and controlling the size of active metals.³ However, the use of noble metals as catalysts,⁴ destruction of the sp² hybridization of the carbon support,⁵ high temperature treatment and long reaction times^4a,6 remain obstacles to wide application of direct alkaline hydrazine fuel cell anodes. Meanwhile, because of the excellent properties of carbon materials such as multi-walled carbon nanotubes (MWCNTs) and graphene in many engineering applications,⁷ the preparation of high-quality carbon-based materials for use as electrocatalysts in fuel cells has been explored in recent years.⁸

In this paper, we present a facile, environmentally friendly, noble metal-free pulse reversal plating method (detailed information is presented in the supporting information†) for the arge scale production of Ni–Fe nanoparticle alloys using the inexpensive materials NiSO₄·6H₂O and FeSO₄·7H₂O as reagents. The process of Ni–Fe nanoparticles deposited on the carbon material is shown in Scheme 1. This approach can completely avoid damaging the sp² hybridization of the carbon materials, allowing the catalyst to exhibit high activity.

Scheme 1 Schematic representation of the experimental procedure for plating Ni–Fe alloy on MWCNTs and graphene.

Carbon-supported Ni–Fe alloy catalysts with stoichiometries from (95 : 5) to (50 : 50) were obtained by pulse reversal plating to determine the optimum content for hydrazine electrocatalysis. Briefly, the carbon material (10 mg) was combined with deionized water (5.0 mL), ethanol (2.5 mL) and Nafion solution (2.5 mL, 0.5 wt%) and then sonicated for 20 min. The ink (10 μL) was applied onto a glassy carbon electrode, left to dry for 10 min under an infrared light and then immersed in a plating solution for 2 min before plating. Ni–Fe alloy catalysts were prepared by pulse reversal plating using a method nearly identical to those previously published.⁹ The compositions of the plating solutions and conditions¹⁰ are listed in Table S1 and S2.† After electrodeposition of the Ni–Fe alloy on the carbon material, cyclic voltammograms (CVs) of the samples were obtained in Ar-saturated 0.1 M hydrazine hydrate–0.015 M KOH–1 M NaCl solution (60% v/v hydrazine hydrate). The catalyst samples (pure Ni, pure Fe, and compositions from Ni₅₅Fe₄₅ to Ni₉₅Fe₅) were tested from −1.3 V to −0.20 V (vs. RHE) at 20 mV s⁻¹, and then cycled 500 times to test the durability of the electrode under these conditions. The electrocatalytic performance was normalized by the mass activity of the Ni–Fe alloy. The total weight of metal can be calculated using the following two equations

M_NiFe = [(Q/2F) × 58.73] × 31.6%

M_NiFe = [(Q/2F) × 58.73] × 20.1%

where M_NiFe is the actual weight of metal, Q is the theoretical electricity consumption, F is the Faraday constant, 31.6% (MWCNTs) and 20.1% (graphene) are the current efficiencies of Ni₈₅Fe₁₅ electroplating, which can be calculated from the ratio of the actual weight to the theoretical weight of metal during plating, and 58.73 is the molar mass of Ni₈₅Fe₁₅.

When the MWCNTs act as the support, the Ni₈₅Fe₁₅ alloy is more active than other compositions (Fig. 1a). The current of the hydrazine hydrate oxidation peak for Ni₈₅Fe₁₅ was higher than that on a pure Ni electrode. No hydrazine oxidation was observed when pure Fe was used as the electrode. As the iron content was increased, the catalytic activity of nanoparticles first increased and then decreased. This shows that nickel plays the major active role in the oxidation of hydrazine, while iron plays a synergistic one. As the iron content was increased, cooperation effects also increased. However, if the content of iron is too much, there are not enough major catalysts (nickel). Experiments show that Ni₈₅Fe₁₅ is the best catalyst for hydrazine oxidation.

Fig. 1 (a) Evaluation of hydrazine electrooxidation activities of Ni–Fe bimetallic catalysts (MWCNTs-based) compared to Ni and Fe (MWCNTs-based) references. Anodic voltammetric sweeps after 500 cycles between −1.30 V and −0.20 V are shown. Conditions: 0.1 M hydrazine hydrate–0.015 M KOH–1 M NaCl, 60 °C, 20 mV s⁻¹. Inset: mass activities at respective peaks. (b) CVs of MWCNTs-based Ni, MWCNTs-based Fe, MWCNTs-based Ni₈₅Fe₁₅, graphene-based Ni, graphene-based Fe, and graphene-based Ni₈₅Fe₁₅ electrocatalysts in 0.1 M hydrazine hydrate–0.015 M KOH–1 M NaCl solution, 60 °C, 20 mV s⁻¹. The metal mass-based catalytic current is plotted.

Fig. 1b compares the CV scans obtained for the 500th cycle of the oxidation of hydrazine hydrate by the series of MWCNT-based (Ni, Fe, Ni₈₅Fe₁₅) and graphene-based (Ni, Fe, Ni₈₅Fe₁₅) catalysts. The figure shows that with either MWCNT or graphene as the support, nickel is the major active catalyst in the oxidation of hydrazine hydrate, while iron is the synergistic one. The catalytic activity of MWCNT-based Ni–Fe nanoparticles is higher than the graphene-based one, and the cause of this phenomenon needs further exploration. The onset oxidation of hydrazine hydrate on Ni–Fe alloy is around −0.8 V and a large oxidation peak was observed at −0.45 V. The high catalytic activity of MWCNT- and graphene-based Ni–Fe nanoparticles is a result of their high surface area, and thorough dispersion on the carbon electrode. In addition, the pulse reversal plating method does not reduce the conductivity of the carbon material because it does not destroy the sp² hybridization of the carbon material (Fig. 3b).

TEM images of graphene- and MWCNT-based catalysts are shown in Fig. 2. The Ni–Fe nanoparticles possessed a size distribution between 10 and 15 nm. Fig. 3a shows XRD patterns of MWCNT-based and graphene-based Ni–Fe nanoparticle samples. Both patterns have three peaks (nearly 43.2°, 52.2° and 75.5°) that appear, indicating all Ni–Fe nanoparticles were face-centered cubic (space group Fm-3m) substitutional solid solutions (disordered alloys). The compositions of the most active catalysts derived from inductively coupled plasma spectra (ICP) are summarized in Table S3,† showing that their actual and nominal compositions are consistent.

Fig. 2 TEM images of (a), and (b) MWCNT-based Ni–Fe nanoparticles, and (c), and (d) graphene-based Ni–Fe nanoparticles.

Fig. 3b shows the Raman spectra of a single-layer of graphene and MWCNT samples before and after plating. After the plating, the D band and G band did not significantly change in either its width or shape. As the Ni–Fe nanoparticle was plated on the carbon material, the I(D) : I(G) ratio increases monotonically from a low of 1.0 ± 0.1 for graphene to 1.1 ± 0.1 for graphene-based Ni–Fe nanoparticles, the I(D) : I(G) ratio of MWCNTs was maintained at 1.1 ± 0.1. This suggests that the sp² carbon network of graphene and MWCNTs was not damaged by plating, so the carbon support should retain its good electrical conductivity.

Fig. 3 (a) XRD pattern of Ni–Fe alloy (MWCNTs-based and graphene-based), (b) Raman spectra of MWCNTs and graphene before and after plating, (c) Tafel curves of pure Ni and Ni₈₅Fe₁₅ alloy in 3.5 wt% NaCl electrolyte, (d) Tafel curves of pure Ni and Ni₈₅Fe₁₅ alloy in 0.7 M NaCl–0.015 M NaOH electrolyte.

The Tafel behavior of the pure Ni and Ni₈₅Fe₁₅ alloy in 3.5 wt% NaCl and 0.7 M NaCl–0.015 M NaOH electrolyte is shown in Fig. 3c and d, respectively. In 3.5 wt% NaCl, E_corr/V of Ni₈₅Fe₁₅ is lower than that of pure Ni, but in 0.7 M NaCl–0.015 M NaOH, E_corr/V of Ni₈₅Fe₁₅ is higher than that of pure Ni. This indicates that the Ni₈₅Fe₁₅ alloy and pure Ni have the same corrosion resistance.

The following mechanism for the Ni₈₅Fe₁₅ nanoparticles with superior electrocatalytic activity towards hydrazine electrooxidation is proposed, see Scheme 2. Initially, the two N atoms of N₂H₄ coordinate with a Ni₈₅Fe₁₅ nanoparticle to form N(ad). As the potential gradually changes from positive to negative, the surface of the Ni₈₅Fe₁₅ nanoparticle becomes electron-rich, which repels nitrogen atoms and attracts hydrogen atoms. Each of the two H–N bonds breaks, so H becomes adsorbed on the metal surface, forming H(ad). Subsequently, H(ad) reacts with OH⁻ to form H₂O and an electron. Finally, N₂ forms between the two residual N(ad) of the original hydrazine molecule. Throughout the process, nickel is the major active catalyst, which leads to the oxidation of hydrazine. As a synergistic dopant, iron significantly promotes the catalytic effect. With iron added to the nickel catalysts, the number of active sites increased, and this feature might contribute to enhance the catalytic activity.

In conclusion, Ni₈₅Fe₁₅ nanoparticles were plated on MWCNTs and graphene using a pulse reversal plating method. These carbon material-based Ni₈₅Fe₁₅ alloy catalysts possess superior electrocatalytic activity for hydrazine electrooxidation, making them attractive for use as direct alkaline hydrazine fuel cell anodes. The fabrication of these catalysts is simple, fast, and does not affect the electrical conductivity of the carbon support. In our experiment, the cost of the catalyst was minimized by using 15 wt% Fe. The corrosion resistance of the Ni₈₅Fe₁₅ alloy was high, making it stable for long term storage and transport.

This research is supported by the Natural Science Foundation of Gansu (No. 1010RJZA218) and the Fundamental Research Funds for the Central Universities (lzujbky-2010-33).

Scheme 2 Schematic representation of hydrazine oxidation.

References

1. J. Sanabria-Chinchilla, K. Asazawa, T. Sakamoto, K. Yamada, H. Tanaka and P. Strasser, J. Am. Chem. Soc., 2011, 133, 5425–5431.
2. T. Sakamoto, K. Asazawa, K. Yamada and H. Tanaka, Catal. Today, 2011, 164, 181–185.
3. G. W. Yang, G. Y. Gao, C. Wang, C. L. Xu and H. L. Li, Carbon, 2008, 46, 747–752.
4. (a) Q. F. Yi and W. Q. Yu, J. Electroanal. Chem., 2009, 633, 159–164; (b) K. Yamada, K. Yasuda, N. Fujiwara, Z. Siroma, H. Tanaka, Y. Miyazaki and T. Kobayashi, Electrochem. Commun., 2003, 5, 892–896; (c) Y. Liang, Y. Zhou, J. Ma, J. Y. Zhao, Y. Chen, Y. W. Tang and T. H. Lu, Appl. Catal., B, 2011, 103, 388–396.
5. D. J. Guo and H. L. Li, Electrochem. Commun., 2004, 6, 999–1003.
6. B. Dong, B. L. He, Y. M. Chai and C. G. Liu, Mater. Chem. Phys., 2010, 120, 404–408.
7. (a) M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110, 132–145; (b) D. Eder, Chem. Rev., 2010, 110, 1348–1385; (c) L. B. Hu, D. S. Hecht and G. Gruner, Chem. Rev., 2010, 110, 5790–5844; (d) N. Karousis, N. Tagmatarchis and D. Tasis, Chem. Rev., 2010, 110, 5366–5397.
8. (a) G. Y. Gao, D. J. Guo, C. Wang and H. L. Li, Electrochem. Commun., 2007, 9, 1582–1586; (b) M. Pumera and H. Iwai, J. Phys. Chem. C, 2009, 113, 4401–4405.
9. (a) A. Bai and C. C. Hu, Electrochim. Acta, 2005, 50, 1335–1345; (b) J. Y. Fei and G. D. Wilcox, Electrochim. Acta, 2005, 50, 2693–2698; (c) P. T. Tang, Electrochim. Acta, 2001, 47, 61–66.
10. (a) T. E. Buchheit, S. H. Goods, P. G. Kotula and P. F. Hlava, Mater. Sci. Eng., A, 2006, 432, 149–157; (b) S. H. Kim, H. J. Sohn, Y. C. Joo, Y. W. Kim, T. H. Yim, H. Y. Lee and T. Kang, Surf. Coat. Technol., 2005, 199, 43–48.

---

Footnote

† Electronic Supplementary Information (ESI) available: The detail of pulse reversal plating. Nickel iron plating chemistry and bath conditions. ICP test data. See DOI: 10.1039/c2ra01346k/

---