An array of leaf-like Co₃Ni microstructures with ferromagnetic properties, superhydrophobic properties and high catalytic performance in the hydrolysis of ammonia borane

Abstract

In this work, a Cu foil supported array of leaf-like Co₃Ni microstructures composed of well aligned nanorods was synthesized by a facile electroless process. It was found that the as-prepared Co₃Ni array exhibited ferromagnetic properties with enhanced coercivity. The static contact angle of water on the surface of the Co₃Ni array could reach 154.7°, indicating that the Co₃Ni array displayed superhydrophobic properties. In addition, the as-prepared Co₃Ni array showed high catalytic activity in the hydrolysis of ammonia borane for hydrogen generation. More importantly, the Co₃Ni array catalyst still retained ca. 91% of its original catalytic activity after 6 cycles, exhibiting significantly improved recyclability and reusability in contrast to many nanocatalysts reported in the literature. The high catalytic performance, together with its ferromagnetic and superhydrophobic properties, makes the Co₃Ni array a multifunctional material with wide applications in different areas.

---

Introduction

The worldwide energy consumption is sharply increasing and the current energy system based on fossil fuels is considered to be a major cause for global climate changes and environmental pollution. So, the development of sustainable, environment-friendly and cost-effective energy sources is undoubtedly the most important task in the 21st century. Owing to its high energy density, zero emission and as an abundant source, hydrogen is regarded as an ideal next-generation fuel.^1,2 Many techniques have been developed to produce and store hydrogen in the past few decades. In particular, generation of hydrogen via catalytic hydrolysis of chemical hydrides, such as NaBH₄, N₂H₄ and NH₃BH₃ (ammonia borane, AB), has been extensively investigated in view of its high efficiency, convenience and safety.^3–5 Among these chemical hydrides, AB has the highest hydrogen storage capacity (19.6 wt%), which renders it more suitable for the application in the production/storage of hydrogen. In the catalytic hydrolysis of AB, catalyst is a key factor which will significantly affect the efficiency of hydrogen generation. In addition to noble metal catalysts, some low-cost catalysts, such as Co,^6,7 Co–B,⁸ Co–Mo–B–P,⁹ Ni,¹⁰ also show relatively high catalytic activity in the AB hydrolysis. However, these catalysts are usually unsupported nanoparticles or nanoparticles supported on powdery substrate, such as activated carbon, Al₂O₃, zeolite.^11–13 Obviously, those catalysts in the form of fine powder are prone to aggregate under the drive of high surface energy, which will reduce their active surface area and thus lower their catalytic activity.¹⁴ Additionally, the tedious and costly separation and recovery procedure of the powdery nanocatalysts is another challenge when these catalysts are in industry-scale application. So, from a technological and operational point of view, it is highly expected to design low-cost catalysts with high stability, good recyclability, as well as high catalytic activity. On the other hand, hierarchical nano/microstructures have attracted increasing attention in recent years because of their excellent performance, enabled by their unique geometry and efficient charge transfer.^15,16 Generally speaking, hierarchical nano/microstructures are composed of numerous nanosized subunits, which are arranged in a specific manner. In the field of catalysis, it has been well verified in literature that the hierarchical nanostructures/microstructures exhibit high catalytic performance in many cases due to their special architectures.^17,18 However, hydrolysis of AB for hydrogen generation catalyzed by hierarchical nanostructures/microstructures has been seldom concerned.

In this work, an array of hierarchical leaf-like Co₃Ni microstructures supported on the Cu substrate was successfully prepared by a facile electroless approach. Compared with the traditional electro-deposition method for fabricating dendritic arrays of metals/alloys of iron group,^19,20 the synthetic process presented in this work is simple and cost-effective. The catalytic performance of the as-prepared Co₃Ni array in the hydrolysis of AB is tested. As far as we know, an alloy of Co and Ni with similar microstructures in the form of an array has not been previously report. Meanwhile, this is the first report on the AB hydrolysis for hydrogen generation catalyzed by hierarchical leaf-like Co₃Ni microstructures. Additionally, owing to its special microstructures, the as-prepared Co₃Ni array exhibits both ferromagnetic property and superhydrophobic property, making it a multifunctional material with wide applications in different fields.

Experimental section

Materials and synthesis

All reagents were of analytic grade, and double-distilled water was used throughout the experiments. To prepare the array of leaf-like Co₃Ni microstructures supported on Cu substrate, 1.5 mmol CoCl₂·6H₂O and 0.5 mmol NiCl₂·6H₂O was dissolved into 20 mL water under intense stirring, followed by the addition of 8 mmol triethylenetetramine and 2 mmol sodium dodecylsulfate to the mixed solution. Then, 10 mL NaOH (20 M), 10 mL hydrazine hydrate solution (50%) and 40 mL ethanol were dropped in sequence to the above solution. Afterwards, the mixture was quickly transferred into a Teflon-lined steel autoclave, in which a piece of 5 cm × 12 cm Cu foil with a thickness of ca. 10 μm and a weight of 0.54 ± 0.02 g (obtained from Guangzhou Great Power Energy &Technology CO., Ltd.) closely attached to the inner wall of vessel was used as support. Then the autoclave was put in the center of a circular column magnet with the magnetic field intensity of ca. 0.2 Tesla around the inner wall of the vessel. Finally, the autoclave, together with the magnet, was kept in an air oven at 120 °C for 8 h.

Characterization

X-Ray diffraction (XRD) patterns were recorded using a PANalytical B. V. Empyean X-ray diffractometer with Cu Kα radiation (λ = 1.5406 Å). The morphology of the array was investigated using a Hitachi S-4800 field emission scanning electron microscope (FE-SEM) equipped with an X-ray energy-dispersive spectrometer (EDS). Transmission electron microscopy (TEM) images were obtained on a JEOL JEM-2100 transmission electron microscope operated at 200 kV. The mass of Co and Ni deposited on the Cu foil was measured by a Varian 720 Inductively Coupled Plasma-Optical Emission Spectrometer (ICP-OES). The magnetic property of the sample was measured by a Quantum Design MPMS XL-7 superconducting quantum interference device magnetometer (SQUID). Contact angle was measured by a contact angle meter (OCA20, DataPhysics Instruments GmbH, Germany) at ambient temperature. The specific surface area was measured on a Quantachrome Autosorb-1 volumetric analyzer, using nitrogen adsorption and the Brunauer–Emmett–Teller method.

Catalytic performance testing

The catalytic performance of the array of leaf-like Co₃Ni microstructures supported on Cu foil was assessed by determining the accumulative volume of hydrogen generated during AB hydrolysis reaction in a glass reactor connected with a gas burette. In a typical procedure, 20 mL of freshly-prepared AB solution (0.2 M) was added in the reactor, followed by the addition of the Co₃Ni array supported on of Cu foil (5 cm × 3 cm). The reaction temperature was fixed at 308 ± 0.5 K in a thermostated reactor. The reusability of the Co₃Ni array was evaluated by repeating the hydrolysis reaction 6 times in such a model: (1) recollecting the catalyst after reaction; (2) washing the catalyst with water and ethanol alternatively before drying at 50 °C in vacuum oven; (3) starting a new run of AB hydrolysis by mixing the catalyst with freshly-prepared AB solution.

Results and discussion

The crystal structures of the Cu substrate, as well as the Cu foil supported Co₃Ni array, are investigated by XRD and the results are shown in Fig. 1. It is observed that bare Cu substrate shows three characteristic peaks at 2θ = 43.4°, 50.6° and 74.2°, corresponding to the (111), (200) and (220) planes of face-centered cubic (fcc) phase of Cu (PDF#89-2838). Apart from these peaks, other three peaks at 2θ = 44.4°, 51.6° and 76.1° are observed in the XRD pattern of the Cu foil supported Co₃Ni array, which can be indexed to the diffractions from the (111), (200) and (220) planes of the fcc phase of CoNi alloy.²¹

Fig. 1 XRD patterns of the Cu foil (a) and the Cu foil supported Co₃Ni array (b).

Fig. 2a and b display the digital camera photos of the bare Cu foil and the Cu foil supported Co₃Ni array, respectively. As can be seen, the Cu foil is originally purple yellow in color and changes into gray black after the deposition of Co₃Ni. The morphology of the Cu supported Co₃Ni array, as well as the Cu foil, is analyzed by FE-SEM and the results are shown in Fig. 2c–f. Clearly, the bare Cu foil is relatively smooth on its surface (Fig. 2c). After the deposition of Co₃Ni, the surface becomes very coarse (Fig. 2d). Fig. 2e indicates that the Co₃Ni layer is composed of numerous leaf-like Co₃Ni microstructures, which are perpendicular to the Cu substrate. The high magnification SEM image in Fig. 2f displays the architecture of a piece of Co₃Ni leaf, which is consisting of a main stem with a diameter of 80–200 nm and many parallel rod-like branches with a diameter of ca. 150 nm on both sides of the stem. The TEM image in Fig. 2g further confirms the leaf-like structure of Co₃Ni in the array. It should be mentioned that even a 30 min ultrasonic treatment can not detach these Co₃Ni microstructures from the Cu substrate, implying that the as-prepared Co₃Ni array is very stable. The typical EDS pattern of a leaf-like Co₃Ni microstructure in Fig. 2h indicates the atomic ratio of Co to Ni is ca. 72 : 28. The ICP-OES analysis demonstrates the atomic ratio of Co to Ni is ca. 74 : 26. Both ratios are very close to the targeted composition of 3 : 1. According to the ICP-OES result, the mass of the Co₃Ni array deposited on Cu foil (5 × 12 cm) is ca. 36.4 mg, and the corresponding catalyst loading on Cu foil is 0.61 mg cm⁻².

Fig. 2 Digital camera photos of the Cu foil (a) and the Cu foil supported Co₃Ni array (b); SEM images of the Cu foil (c) and the Cu foil supported Co₃Ni array (d–f); TEM image (g) and EDS pattern (h) of the Co₃Ni array.

In this study, Co₃Ni alloy was produced by the co-reduction of Co²⁺ and Ni²⁺. At the early stage of the reduction process, plenteous primary Co₃Ni alloy nuclei are formed in solution, which will be magnetized and migrate to the Cu substrate under the magnetic force provided by the external magnet. After they are deposited on the Cu foil, these nuclei will act as crystal nuclei and provide heterogeneous nucleation sites for crystal growth. As the reduction reaction proceeds, more and more Co₃Ni clusters will be formed and move to the Cu substrate, then add themselves to existing nuclei. In such model, nanostructures on the Cu substrate are formed.

To determine the magnetic properties of the Cu supported array of leaf-like Co₃Ni microstructures, magnetic measurements at 298 K were carried out on the array, with the applied field parallel and vertical to the Cu substrate. The magnetic hysteresis loops are displayed in Fig. 3, which demonstrate the array has typical ferromagnetic properties. When the applied external magnetic was parallel to the Cu substrate, the saturation magnetization, the remnant magnetization and the coercivity of the array is 114.0 emu g⁻¹, 58.8 emu g⁻¹ and 79.5 Oe, respectively. While when the applied external magnetic was vertical to the Cu substrate, the saturation magnetization, the remnant magnetization and the coercivity of the array is 96.7 emu g⁻¹, 11.1 emu g⁻¹ and 286.8 Oe, respectively. Compared with the coercivity of the nanosized Co₇₅Ni₂₅ alloy (31.69 Oe),²² our Co₃Ni array exhibits much higher coercivity. It has been demonstrated that a nonspherical nanosized magnetic materials can possess shape anisotropy and a small departure from sphericity in shape will cause a significant increase of coercivity.²³ In the present study, the shape anisotropy of the leaf-like Co₃Ni microstructures may be responsible for the improved coercivity.

Fig. 3 Magnetic hysteresis loops of the Cu foil supported Co₃Ni array with the applied external magnetic field parallel and vertical to the Cu substrate.

To evaluate the wettability of the Co₃Ni array, the static contact angle was determined. Fig. 4 is the typical photograph of a water droplet on the Cu supported Co₃Ni array. The contact angle of the Co₃Ni array is 154.7°, indicating that the Co₃Ni array exhibit superhydrophobic property. It is known that the superhydrophobic materials can be obtained by the construction of appropriate surface hierarchical micro/nanostructures and chemical composition.²⁴ As far as our sample concerned, the leaf-like Co₃Ni microstructures in the array consist of well aligned nanorods. Such hierarchical micro/nanostructures on the surface of the Cu substrate account for the superhydrophobic properties of the sample.

Fig. 4 Typical photograph of a water droplet on the surface of Cu foil supported Co₃Ni array.

In the present study, the as-prepared Cu foil supported array of leaf-like Co₃Ni microstructures as a catalyst for the hydrolysis of AB is investigated. To clarify the role of Cu substrate, hydrolysis of AB in the presence of bare Cu foil is carried out. It is found that nearly no hydrogen is generated, suggesting that the Cu foil in our work has no catalytic activity in the hydrolysis reaction. In contrast, when the array of leaf-like Co₃Ni microstructures acts as a catalyst, hydrogen is continuously generated. Fig. 5 shows the relationship of the accumulated volume of hydrogen vs. reaction time at different catalyst dosages in terms of geometrical surface areas of film catalyst, which are the product of the length and the width of the film. As can be observed, the accumulated volume of hydrogen is directly proportional to the reaction time, hinting that the hydrolysis of AB catalyzed by the Co₃Ni array is zero-order with respect to the concentration of AB. This implies that the adsorption of AB on the array surface may be a rate-limiting step.²⁵ Noting that the volume of hydrogen at the late stage of the hydrolysis (after 20 min) is smaller than the normal value at catalyst dosages of 22.5 and 30.0 cm². Similar results have also been reported by other groups.^8,10 The corresponding rate constants can be calculated from the slopes of fitting lines in Fig. 5 and the results are listed in Table 1. The total turnover frequency (TOF) of catalyst can also be calculated based on these data. At catalyst dosage of 9.1 mg, the value of TOF is ca. 2.03 mol H₂ min⁻¹ (mol cat.)⁻¹.

Fig. 5 Accumulative volume of hydrogen generated during AB hydrolysis vs. reaction times at different catalyst dosages in terms of geometrical surface (T = 308 K).

Table 1 Comparison of rate constants and BET surface normalized rate constants of AB hydrolysis when Co₃Ni array and Co₃Ni nanoparticles act as catalysts

Catalyst dosage	Rate constant (mL min^−1)		BET surface area normalized rate constant (mL min^−1 m^2)
	Co3Ni array	Co3Ni nanoparticles	Co3Ni array	Co3Ni nanoparticles
4.6 mg	4.5	3.3	72.5	32.8
9.1 mg	7.4	4.8	60.2	24.1
13.7 mg	10.5	6.6	56.8	22.0
18.4 mg	11.7	7.2	47.1	17.9

---

It is worth noting that production of hydrogen via AB hydrolysis catalyzed by various catalysts has been widely reported and a huge number of data on the rate constants of AB hydrolysis have been published in literature. However, the detailed reaction conditions of AB hydrolysis, such as reaction model, reaction temperature, catalyst dosage, concentration of AB and even the support, are not the same in different reports. So, it may be unsound to assess these catalysts by comparing their rate constants or TOF without normalizing these parameters. In this study, for comparison, Co₃Ni nanoparticles with a diameter of 40–50 nm synthesized by a modified polyol method²⁶ instead of the leaf-like Co₃Ni array were used as a catalyst while other reaction conditions were kept the same. The corresponding results are also shown in Table 1. As can be seen, our Co₃Ni array exhibit higher catalytic activity than Co₃Ni nanoparticles in the hydrolysis of the AB. Two possible reasons account for the fact that the Co₃Ni array shows much higher catalytic activity than Co₃Ni nanoparticles. Firstly, Co₃Ni layers supported on substrate are composed of shape-anisotropic leaf-like microstructures, which have more corners and edges than the spherical nanoparticles. Generally, the atoms on the corners or edges have unsaturated valency with less number of bonds around them than those in the interiors or on the faces.²⁷ Such atoms have much higher intrinsic catalytic activity. In this study, the BET surface areas of Co₃Ni array and the Co₃Ni nanoparticles were measured, which were 13.5 and 21.9 m² g⁻¹, respectively. Correspondingly, the BET surface area normalized rate constants can be figured out. As shown in Table 1, BET surface area normalized rate constants of Co₃Ni array is significantly larger than those of Co₃Ni nanoparticles, hinting that Co₃Ni array has higher intrinsic catalytic activity than Co₃Ni nanoparticles. Secondly, these Co₃Ni nanoparticles suspended in AB solution during hydrolysis reaction tend to agglomerate due to their high surface energy, as well as the magnetic attraction among the Co₃Ni nanoparticles resulted from their ferromagnetic property.¹⁴ In this case, the surface area of particulate Co₃Ni catalyst will decrease, resulting in the decline of catalytic performance. In contrast, the leaf-like Co₃Ni microstructures in the array are immobilized on the surface of substrate and therefore the decrease of catalytic activity related to the aggregation is evitable.

The apparent activated energy (E_a) is a very important parameter of a catalyst, which can be applied to evaluate its catalytic performance. In general, a catalyst with lower activation energy in the catalytic reaction always shows a higher catalytic activity. In this study, the hydrolysis of AB catalyzed by the Co₃Ni array was carried out at various temperatures to obtain the E_a value. As displayed in Fig. 6a, the hydrolysis of AB is enhanced as the reaction temperature increases. The hydrolysis rate is 5.0, 7.4, 10.5 and 19.4 mL min⁻¹ at reaction temperature of 298 K, 308 K, 318 K and 328 K, respectively. On the basis of these data, E_a of the AB hydrolysis catalyzed by the leaf-like Co₃Ni array can be calculated according to the classical Arrhenius equation, which is 35.6 kJ mol⁻¹ (Fig. 6b). This value is higher than that of Co₃₅Pd₆₅/C catalyst (27.5 kJ mol⁻¹)²⁵ and Co–P catalyst (22 kJ mol⁻¹),²⁸ but is significantly lower than those of recently reported Co-based and Ni-based catalysts, such as Co/γ-Al₂O₃ (62 kJ mol⁻¹),¹¹ Co/zeolite (56 kJ mol⁻¹),¹³ PVP–Co (46 kJ mol⁻¹)²⁹ and Co_0.75B_0.25 (40.85 kJ mol⁻¹),³⁰ bulk Ni (70 kJ mol⁻¹),³¹ CoNi/RGO (39.89 kJ mol⁻¹)³² and Ni_0.97Pt_0.03 (57 kJ mol⁻¹).³³

Fig. 6 (a) Accumulative volume of hydrogen generated during AB hydrolysis vs. reaction times at different reaction temperatures; (b) the logarithm of rate constants vs. reciprocal of reaction temperature.

The reusability and durability of a catalyst are important concerns to be considered before its practical applications. In this work, the normalized rate constant, i.e. k_n/k₁, where k₁ and k_n represents respectively the rate constant of AB hydrolysis in cycle 1 and cycle n, is applied to assess the durability and reusability of the as-prepared array of leaf-like Co₃Ni microstructures supported on the Cu foil. As can be observed in Fig. 7, the normalized rate constant is as high as 0.91 after 6 cycles, indicating that the used catalyst still retains ca. 91% catalytic activity of its original activity. In contrast, the Co₃Ni nanoparticles obtained via the modified polyol method lose ca. 40% activity after 6 cycles. Moreover, the as-prepared Cu supported array of leaf-like Co₃Ni microstructures also exhibits much better durability and reusability than many Co or Co-containing nanocatalysts in literature, such as intrazeolite Co nanoclusters (31% activity loss in the fifth run),¹³ CoNi/RGO (31% activity loss in the fifth run).³² For those Co or Co-containing nanocatalysts in the form of powder, the material loss during the process of separation, drying and re-dispersed is a common reason that causes the decrease in the activity of the catalysts.¹³ Also, under the drive of high surface energy and the attraction of magnetic force among each other, the fine powdery particles are very likely to aggregate during the process of usage, isolation and desiccation, which will result in a remarkable loss of the catalytic activity in the long-term run. In contrast, the leaf-like Co₃Ni microstructures in the array are immobilized on the Cu foil, so the activity drop caused by the migration and aggregation of Co₃Ni microstructures is evitable. Besides, the alteration of shapes and morphology of micro- or nanocatalysts during catalytic process can also influence their catalytic performance.³⁴ To clarify this, the SEM image of the Co₃Ni array catalyst after 6 cycles are shown in Fig. 8, demonstrating that there is no remarkable morphology change in the Co₃Ni array. This observations hint that the Co₃Ni array is stable under our experiment conditions.

Fig. 7 Normalized rate constant at different cycles.

Fig. 8 SEM image of the Co₃Ni array after 6 cycles.

Conclusions

In summary, an array of leaf-like Co₃Ni microstructures supported on the Cu foil was prepared, which showed high catalytic activity in the hydrolysis of AB. The apparent activated energy of the AB catalyzed by the as-prepared Co₃Ni array was calculated to be ca. 35.6 kJ mol⁻¹, which is much lower than those of many Co-based and Ni-based nanocatalyst in literature. Also, it displayed better durability and reusability in contrast to many powdery Co-based nanocatalysts. Owing to its high catalytic activity, good durability and reusability, as well as the low cost, the array of leaf-like Co₃Ni microstructures showed great potential in the applications in the hydrolysis of ammonia borane for hydrogen generation. In addition, the as-prepared Co₃Ni array exhibits both ferromagnetic property and superhydrophobic property due to its special architecture, which make it a multifunctional material with wide applications in different fields.

Acknowledgements

This work was financially supported by the Science and Technology Commission of Shanghai Municipality (No. 14DZ2261000), the National Natural Science Foundation of China (No. 51001052), the High-level Talent Project of the University in Guangdong Province (184), the Distinguished Young teachers in Higher Education of Guangdong Province (No. Yq2013154), and the Characteristic and Innovative Project of the Education Department of Guangdong Province (No. 2014KTSCX177).

Notes and references

1. J. A. Turner, Science, 2004, 305, 972–974.
2. Z.-M. Huang, A. Su and Y.-C. Liu, Energy, 2013, 51, 230–236.
3. G. H. Liu and Z. P. Li, J. Power Sources, 2009, 187, 527–534.
4. W. Chen, J. Ji, X. Duan, G. Qiang, P. Li, X. Zhou, D. Chen and W. Yuan, Chem. Commun., 2014, 50, 2142–2144.
5. L. He, Y. Huang, X. Y. Liu, L. Li, A. Wang, X. Wang, C.-Y. Mou and T. Zhang, Appl. Catal., B, 2014, 147, 779–788.
6. M. Paladini, G. M. Arzac, V. Godinho, M. C. Jiménez De Haro and A. Fernández, Appl. Catal., B, 2014, 158–159, 400–409.
7. J. Liao, H. Li and X. Zhang, Catal. Commun., 2015, 67, 1–5.
8. N. Patel, R. Fernandes, G. Guella and A. Miotello, Appl. Catal., B, 2010, 95, 137–143.
9. R. Fernandes, N. Patel, A. Miotello and L. Calliari, Top. Catal., 2012, 55, 1032–1039.
10. M. Mahyari and A. Shaabani, J. Mater. Chem. A, 2014, 2, 16652–16659.
11. Q. Xu and M. A. Chandra, J. Alloys Compd., 2007, 446–447, 729–732.
12. Q. Xu and M. Chandra, J. Power Sources, 2006, 163, 36–370.
13. M. Rakap and Ö. Saim, Int. J. Hydrogen Energy, 2010, 35, 3341–3346.
14. Y. Du, H. Chen, R. Chen and N. Xu, Appl. Catal., A, 2004, 277, 259–264.
15. J. Bai, X. Li, G. Liu, Y. Qian and S. Xiong, Adv. Funct. Mater., 2014, 24, 3012–3020.
16. M. U. Anu Prathap, S. Sun, C. Wei and Z. J. Xu, Chem. Commun., 2015, 51, 4376–4379.
17. C. Deng, X. Ge, H. Hu, L. Yao, C. Han and D. Zhao, CrystEngComm, 2014, 16, 2738–2745.
18. M. Wang, T. Sun, Y. Shi, G. Jiang and Y. Tang, CrystEngComm, 2014, 16, 10624–10630.
19. K. H. Kim, J. Y. Zheng, W. Shin and Y. S. Kang, RSC Adv., 2014, 2, 4759–4767.
20. J. Y. Zheng, Z. L. Quan, G. Song, C. W. Kim, H. G. Cha, T. W. Kim, W. Shin, K. J. Lee, M. H. Jung and Y. S. Kang, J. Mater. Chem., 2012, 22, 12296–12304.
21. H. Li, J. Liao, Y. Feng, S. Yu, X. Zhang and Z. Jin, CrystEngComm, 2012, 14, 2974–2980.
22. M. Shen, Y.-F. Wang, F. Zhang and Q.-S. Wu, J. Phys. Chem. C, 2009, 113, 5960–5966.
23. D. L. Leslie-Pelecky and R. D. Rieke, Chem. Mater., 1996, 8, 1770–1783.
24. K. Liu and L. Jiang, Nanoscale, 2011, 3, 825–838.
25. D. Sun, V. Mazumder, Ö. Metin and S. Sun, ACS Nano, 2011, 5, 6458–6464.
26. J. Bregado-Gutiérrez, A. J. Saldívar-García and H. F. Lopez, Mater. Lett., 2008, 62, 939–942.
27. M. A. Mahmoud, F. Saira and M. A. El-Sayed, Nano Lett., 2010, 10, 3764–3769.
28. K. S. Eom, K. W. Cho and H. S. Kwon, Int. J. Hydrogen Energy, 2010, 35, 181–186.
29. Ö. Metin and S. Özkar, Energy Fuels, 2009, 23, 3517–3526.
30. A. K. Figen, Int. J. Hydrogen Energy, 2013, 38, 9186–9197.
31. S. B. Kalidindi, M. Indirani and B. R. Jagirdar, Inorg. Chem., 2008, 47, 7424–7429.
32. Y. Yang, F. Zhang, H. Wang, Q. Yao, X. Chen and Z.-H. Lu, J. Nanomater., 2014, 2014, 1–9.
33. F. Cheng, H. Ma, Y. Li and J. Chen, Inorg. Chem., 2007, 43, 788–794.
34. X. Liu, D. Wang and Y. Li, Nano Today, 2012, 7, 448–466.

---