Synergistically thermodynamic and kinetic tailoring of the hydrogen desorption properties of MgH₂ by co-addition of AlH₃ and CeF₃

Abstract

MgH₂ possesses a high hydrogen capacity and excellent reversibility. However, the high thermal stability and slow sorption kinetics retard its practical application as an on-board hydrogen storage material. In this work, AlH₃ and CeF₃ were introduced into Mg-based materials for the purpose of improving both the thermodynamic and the kinetic properties of MgH₂. DSC-TG analysis shows that the onset hydrogen desorption temperature of MgH₂ can be synergistically reduced by 86 °C through the co-addition of 0.25AlH₃ and 0.01CeF₃. Isothermal desorption measurements demonstrate that the co-addition of AlH₃ and CeF₃ significantly enhances the hydrogen desorption kinetics of MgH₂ with the absence of the induction period in the initial stage and the acceleration of the hydrogen desorption process. In addition, this co-doped MgH₂ shows very good cycling stability at 300 °C with a 1 h capacity of 3.5 wt% and a 3 h capacity of 4.5 wt%. Structural analysis by XRD measurements indicates that during the hydrogen desorption process, MgH₂ may react with Al (generated from the in situ decomposition of AlH₃) to form Mg solid solution and Mg₁₇Al₁₂, which contribute to the thermodynamic improvement of the Mg-based material. In addition, MgH₂ may also react with CeF₃ to form MgF₂ and CeH_2–3, which act both as hydrogen diffusion gateways and as an impediment to the grain growth of MgH₂ during hydrogen sorption cycling, thus improving the hydrogen desorption kinetics and the cycling stability of MgH₂. Finally, it was found that the presence of AlH₃ kinetically helps CeF₃ to exert its positive effect on the hydrogen desorption properties of MgH₂. This work provides a method for simultaneously tailoring the thermodynamic and kinetic properties of MgH₂ by the synergistic addition of metal hydride and rare earth fluoride.

---

1 Introduction

Magnesium hydride (MgH₂) is a promising candidate as a hydrogen storage material for on-board applications due to its high hydrogen capacity and excellent reversibility.¹ However, the unfavourably high thermodynamic stability and sluggish hydrogen sorption kinetics restrict the practical application of MgH₂.^2,3

Mixing or alloying MgH₂/Mg with metals is one strategy used to overcome these problems.^4–14 Among the various metals, metallic aluminium (Al) has been studied to improve the hydrogen sorption properties of MgH₂.^5,7,15,16 Zaluska et al.¹⁷ suggested that Al may not only act as a heat-transfer medium, but also be involved in the hydrogen sorption reactions of MgH₂. In our previous work,^18,19 we utilised aluminium hydride (AlH₃) as an Al source in place of metallic Al to significantly reduce the hydrogen desorption temperature of MgH₂. It was demonstrated that in the Mg–Al–H system, AlH₃ is much better than metallic Al in terms of its ability to improve the hydrogen desorption properties of MgH₂. This superiority was ascribed not only to the brittleness of AlH₃, but also to the fact that AlH₃ undergoes decomposition to form oxide-free Al* upon heating. Its brittleness makes it easier for AlH₃ to mix well with MgH₂ to ensure uniform elemental distributions by short-term milling. The oxide-free Al* may benefit the reaction kinetics since it is of high chemical activity. However, the hydrogen desorption kinetics of the Mg–Al–H system suffers from a severe decline when subjected to cycling.²⁰ It was indicated that this kinetic decline is due to grain growth or particle agglomeration during the hydrogen sorption process.

Mixing with transition metal halides is another strategy that has been employed to successfully improve the hydrogen sorption kinetics of Mg-based materials.^21–29 It was suggested that many halides may react with MgH₂, and the resulting products impede the grain growth of MgH₂, thus improving the kinetics of MgH₂. However, while these metal halides can affect the kinetics of MgH₂, they hardly change the thermodynamics of MgH₂. It is true that improved kinetics are essential for fast hydrogen uptake and release; however, the thermodynamic destabilisation of MgH₂ is also highly desired because a lower stability means that MgH₂ is able to start releasing hydrogen at a lower temperature.

In the present work, to achieve the simultaneous enhancement of the thermodynamics and kinetics of MgH₂, AlH₃ as the destabilizing agent and a transition metal fluoride (cerium fluoride, CeF₃) as the accelerating agent were introduced together into Mg-based materials by ball milling. The following results show that AlH₃ can thermodynamically destabilise MgH₂, while CeF₃ can kinetically accelerate the release of hydrogen. In addition, it is interesting that the co-addition of AlH₃ and CeF₃ can synergistically enhance the desorption properties of MgH₂; the co-doped MgH₂ shows fast desorption kinetics and very good cycling stability at 300 °C. Moreover, CeF₃ was reported by Jin et al.²¹ to have little influence on the hydrogen desorption of MgH₂. However, in the present work, CeF₃ was found to significantly reduce the desorption temperature of MgH₂, especially in the presence of a small amount of AlH₃.

2 Experimental sections

Experimentally, MgH₂ (Alfa Aesar, 98%) and CeF₃ (Sinopharm Group, 99.9%) were used as received, and AlH₃ was synthesised by a wet chemical method.^30–32 The samples were then prepared by ball milling the starting materials using a QM-3SP4 planetary mill (Nanjing Nanda Instrument Plant). In general, 1 g of powder and 50 g of stainless steel balls were sealed in a stainless steel vial with a volume of 100 mL. The milling was carried out at 400 rpm for 30 min. All handling was performed under inert atmosphere. Four samples (MgH₂, MgH₂ + 0.25AlH₃, MgH₂ + 0.01CeF₃, and MgH₂ + 0.25AlH₃ + 0.01CeF₃) were prepared for comparison, and two series of samples (MgH₂ + mCeF₃, m = 0.01, 0.02, and 0.05 and MgH₂ + 0.1AlH₃ + nCeF₃, n = 0.01, 0.02, and 0.05) were prepared for further study.

Phase structures were determined by powder X-ray diffraction (XRD) using a PANalytical X-ray diffractometer (X'Pert Pro) with Cu Kα radiation. During sample transfer and measurement, the samples were sealed with amorphous membranes to avoid exposure to air.

Nonisothermal hydrogen desorption was studied by differential scanning calorimetry (DSC) and thermogravimetry (TG) analysis (Netzsch, STA449F3). The samples were heated from room temperature to 500 °C at a rate of 10 °C min⁻¹. During the heating process, argon was flowed at 50 mL min⁻¹ to prevent sample oxidation.

Isothermal desorption was carried out on a lab-built Sieverts-type apparatus. Before the desorption measurement, the sample holder was evacuated under vacuum. The samples were then rapidly heated to 300 °C and held at this temperature for the collection of desorption curves. Prior to the next desorption cycle, the dehydrogenated sample was rehydrogenated at 300 °C and 5 MPa H₂ for 1 h.

Field emission scanning electron microscopy (SEM; FEI SU 70) was used to study the morphologies of the samples.

3 Results and discussion

The XRD patterns of the as-synthesised AlH₃ and the as-milled samples (Fig. S1, ESI†) indicate they are mainly the physical mixtures of the starting materials after milling. No CeF₃-relevant phases are detected in the CeF₃-doped samples, which may due to the low addition content.

Fig. 1 contains the DSC curves of the samples with a heating rate of 10 °C min⁻¹. The DSC curves show that the peak desorption temperature of the as-milled MgH₂ is 388 °C, which is reduced to 375 (or 354 °C) when it is mixed with 0.25AlH₃ (or 0.01CeF₃). A further reduction to 346 °C can be achieved by the co-addition of 0.25AlH₃ and 0.01CeF₃, indicating a synergistic enhancement effect.

Fig. 1 DSC curves of the as-milled MgH₂ (a), MgH₂ + 0.25AlH₃ (b), MgH₂ + 0.01CeF₃ (c), and MgH₂ + 0.25AlH₃ + 0.01CeF₃ (d) samples.

Fig. 2 shows the TG curves of the samples. The weight loss of each sample during heating is indicated in the figure with the theoretical weight loss included in parentheses. The onset desorption temperature of the as-milled MgH₂ is 350 °C, which is reduced to 292 °C (or 301 °C) when it is mixed with 0.25AlH₃ (or 0.01CeF₃). Further reduction of the onset desorption temperature to 264 °C can be achieved by the co-addition of 0.25AlH₃ and 0.01CeF₃. This desorption temperature of the co-doped MgH₂ is 86 °C lower than that of the as-milled MgH₂. Similar to the DSC results, the onset desorption temperature of the co-doped MgH₂ is the lowest among the samples studied. These DSC-TG results jointly demonstrate that AlH₃ and CeF₃ have synergistic effects on the hydrogen desorption temperature of MgH₂.

Fig. 2 TG curves of the as-milled MgH₂ (a), MgH₂ + 0.25AlH₃ (b), MgH₂ + 0.01CeF₃ (c), and MgH₂ + 0.25AlH₃ + 0.01CeF₃ (d) samples.

The hydrogen desorption kinetics of the samples are investigated by isothermal desorption at 300 °C, and the fractional hydrogen desorption curves are displayed in Fig. 3. It is observed that the as-milled MgH₂ releases a very small amount of hydrogen after desorption at 300 °C for 3 h. This can also be confirmed by the XRD pattern of the desorption product (Fig. S2, ESI†), which shows that the majority of MgH₂ is still present in the desorption product. The hydrogen desorption of AlH₃-doped MgH₂ first accelerates in the initial 30 min and then slows in the following stage. As for the CeF₃-doped MgH₂, the hydrogen desorption first undergoes an induction period in the initial 20 min and then proceeds with fast kinetics. However, it is interesting that with the co-addition of AlH₃ and CeF₃, the induction period is absent, and the desorption proceeds rapidly with a speed as high as that of the CeF₃-doped MgH₂. In addition, the extent of hydrogen desorption for the co-doped MgH₂ is over 90%, which is the highest among the samples studied.

Fig. 3 Isothermal hydrogen desorption curves of the as-milled samples plotted as fractional desorption of MgH₂ in the sample vs. time. The desorption temperature is 300 °C.

Fig. S2 (ESI†) shows the XRD patterns of the isothermal desorption products of each sample. The desorption product of the MgH₂ + 0.25AlH₃ sample mainly contains Mg₁₇Al₁₂, Mg, and some un-decomposed MgH₂, while the product of the MgH₂ + 0.01CeF₃ sample mainly contains Mg and some traces of un-decomposed MgH₂. However, no CeF₃-relevant phases are detected; this may be due to the low addition content of CeF₃. As for the MgH₂ + 0.25AlH₃ + 0.01CeF₃ sample, its product mainly contains Mg₁₇Al₁₂ and Mg phases, and no un-decomposed MgH₂ is detected, suggesting that MgH₂ is fully decomposed by the co-addition of 0.25AlH₃ and 0.01CeF₃. A diffraction peak located at 2Theta = 27°–28° is suspected of belonging to CeH_2–3 phase. Although the CeF₃-relevant phases cannot be detected here, the XRD results at least demonstrate that only partial MgH₂ decomposition occurs in the AlH₃-doped and CeF₃-doped MgH₂ samples, but almost full decomposition occurs in MgH₂ co-doped with AlH₃ and CeF₃.

Combining the kinetic results with the structural analysis, we conclude that the co-addition of AlH₃ and CeF₃ synergistically improves the desorption kinetics and the desorption extent of MgH₂.

The cycling desorption properties of MgH₂ co-doped with AlH₃ and CeF₃ were studied by measuring the isothermal desorption curves at 300 °C for 10 cycles (Fig. 4a). As can be seen, the desorption kinetics first accelerates at the second cycle and then is maintained for the following cycles. Fig. 4b shows that the capacity of 1 h desorption first increases in the initial three cycles and then is maintained at a value of ca. 3.6 wt% with the exception of a slight decrease at cycle eight. The capacity of 3 h desorption is mainly maintained at a value of ca. 4.5 wt%. These cycling desorption measurements indicate that the cycling desorption stability of MgH₂ co-doped with AlH₃ and CeF₃ is extremely good.

Fig. 4 (a) Cycling isothermal desorption curves of the MgH₂ + 0.25AlH₃ + 0.01CeF₃ sample. (b) Capacity of hydrogen released from the MgH₂ + 0.25AlH₃ + 0.01CeF₃ sample after desorption for 1 h and 3 h vs. cycle number. The desorption temperature is 300 °C.

It has been shown in Fig. S2 (ESI†) that due to the low addition amount, we cannot detect any CeF₃-relevant phase by XRD in either of the as-milled samples (Fig. S1†) or in the desorption product (Fig. S2 (ESI†)). In order to facilitate the detection of the CeF₃-relevant phases using XRD, the content of CeF₃ was increased, and two series of samples, i.e., MgH₂ + mCeF₃ (m = 0.01, 0.02, and 0.05) and MgH₂ + 0.1AlH₃ + nCeF₃ (n = 0.01, 0.02, and 0.05), were prepared by ball milling. Their XRD patterns are shown in Fig. S3 (ESI†). It can be seen that the as-milled samples are mainly the physical mixtures of the starting materials. CeF₃ is detected in the as-milled samples (see Fig. S3d–i (ESI†)) and does not react with MgH₂ or AlH₃ during ball milling. Scherrer's equation³³ was utilised to estimate the grain sizes of MgH₂ in the samples, which are indicated in the figure. The grains of MgH₂ can be refined from 43–44 nm to 38–40 nm by the addition of 0.01CeF₃, and further refinement to 25–28 nm can be achieved by the addition of 0.05CeF₃. That is to say, CeF₃ addition is likely to refine the grains of MgH₂.

The XRD patterns of the desorption products of MgH₂ + 0.1AlH₃, MgH₂ + 0.05CeF₃, AlH₃ + 0.1CeF₃, and MgH₂ + 0.1AlH₃ + 0.02CeF₃ are presented in Fig. 5. The desorption product of the MgH₂ + 0.1AlH₃ sample contains Mg, Mg₁₇Al₁₂, and un-decomposed MgH₂, which means that MgH₂ reacted with Al to form Mg₁₇Al₁₂. Mg, a small amount of un-decomposed MgH₂, MgF₂, and CeH_2–3 exist in the desorption product of the MgH₂ + 0.05CeF₃ sample; this suggests that CeF₃ may have reacted with MgH₂ during the desorption process to form MgF₂ and CeH_2–3. The desorption product of the AlH₃ + 0.1CeF₃ sample indicates that AlH₃ prefers to decompose by itself and does not react with CeF₃ during the desorption process since CeF₃ remains unreacted in the sample before and after desorption. As for the MgH₂ + 0.1AlH₃ + 0.02CeF₃ sample, a mixture of Mg, Mg₁₇Al₁₂, MgF₂, CeH_2–3, and some traces of un-decomposed MgH₂ exist in the desorption product. These products suggest that on the one hand, Al (formed from the decomposition of AlH₃) reacts with MgH₂ to form Mg₁₇Al₁₂; on the other hand, CeF₃ also reacts with MgH₂ to form MgF₂ and CeH_2–3. From the inset in Fig. 5, it is observed that the diffraction peak of Mg in the desorption products of MgH₂ + 0.1AlH₃ and MgH₂ + 0.1AlH₃ + 0.02CeF₃ has shifted to higher angles, which indicates that some Al atoms have dissolved into the crystal structure of Mg, forming an Al-doped Mg solid solution (denoted as Mg_ss(Al)).

Fig. 5 XRD patterns of the desorption products of the selective samples. The hydrogen desorption measurements were carried out isothermally at 300 °C for 3 h.

Based on the above phase analysis of the desorption products of each sample, we can conclude that after hydrogen desorption, Al (formed from the decomposition of AlH₃) reacted with MgH₂ to form an Mg solid solution (Mg_ss(Al)) and Mg₁₇Al₁₂ according to eqn (1) and (2):

MgH₂ + Al* → Mg_ss(Al) + H₂ (1)

MgH₂ + Al* → Mg₁₇Al₁₂ + H₂ (2)

Meanwhile, CeF₃ reacted with MgH₂ to form MgF₂ and CeH_2–3 following eqn (3). CeH_2–3 is a nonstoichiometric metal hydride.

MgH₂ + CeF₃ → MgF₂ + CeH_2–3 (3)

The formations of Mg solid solution and Mg₁₇Al₁₂ indicate that the thermal stability of MgH₂ may have been reduced by the addition of AlH₃ since several theoretical studies have reported that the stability of MgH₂ can be changed by doping with Al. Shang et al.⁷ used first-principles calculations to study the stability of MgH₂ with metal doping and found that the formation heat of MgH₂ can be reduced from −75.99 kJ mol⁻¹ H₂ for pure MgH₂ to −28.36 kJ mol⁻¹ H₂ for MgH₂ doped with 20 mol% Al. Another theoretical calculation by Kelkar et al.¹⁵ showed a reduction of 9.3 kJ mol⁻¹ H₂ for the heat of formation of MgH₂ with the addition of 6.25 mol% Al. The Mg₁₇Al₁₂ alloy was reported to have better hydrogen desorption/absorption properties than the pure Mg/MgH₂ materials. Bouaricha et al.⁵ prepared Mg₁₇Al₁₂ alloy by ball milling elemental Mg and Al. They found that Mg₁₇Al₁₂ shows improved hydrogen sorption properties compared to pure Mg and can be reversibly hydrided into MgH₂ and Al under conditions of 400 °C and 38 bar H₂. It was ascertained that the hydrogenation of Mg₁₇Al₁₂ proceeds in two steps:^5,34,35 first, Mg₁₇Al₁₂ is hydrided into Mg₂Al₃ and MgH₂; then, Mg₂Al₃ is further hydrided into MgH₂ and Al in the second step. These reactions are totally reversible. Crivello et al.³⁵ demonstrated that the presence of Al in the Mg₁₇Al₁₂ materials leads to an increase of their plateau pressures, which means that Al can destabilise MgH₂ by forming Mg–Al alloys. Therefore, MgH₂ was thermodynamically destabilised by AlH₃ in the present work, and this may have contributed to the absence of the induction period in the isothermal desorption curves of AlH₃-doped MgH₂ in Fig. 3. As the isothermal desorption was carried out by quickly heating the sample to 300 °C (within 10–15 min) followed by holding at this temperature, which is very close to the practical heating condition, the destabilised AlH₃-doped MgH₂ can start to release hydrogen at a lower temperature than MgH₂ without AlH₃ addition, which results in the absence of the induction period.

With respect to the formation of MgF₂, it was suggested that MgF₂ may benefit the initial activation of MgH₂ during the hydrogen desorption process.³⁶ As for CeH_2–3, some studies have reported the formation of this rare earth hydride from either the as-milled MgH₂–Ce composite or the hydrogenation product of Mg–Ni–Ce alloys.^12,27,37,38 However, the CeH_2–3 in the present work was generated from the reaction between MgH₂ and CeF₃. Transition metal fluorides like CeF₃ are generally very fine, which leads to a more uniform distribution of CeH_2–3 on the particle surfaces or the grain boundaries of MgH₂. CeH_2–3 may act as a “hydrogen pump” that helps deliver hydrogen atoms from/to the MgH₂/Mg matrix during the hydrogen desorption–absorption process, thus improving the hydrogen sorption kinetics of MgH₂ (Fig. 3). On the other hand, CeH_2–3/MgF₂ layers can also impeded the grain growth or coarsening of MgH₂ during the cycling sorption process. These two factors are the main cause of the improvement in the desorption kinetics (Fig. 3) and the extremely good cycling stability (Fig. 4) of MgH₂.

A schematic diagram displaying the structural evolutions of AlH₃ and CeF₃ during the sample preparation and hydrogen desorption processes can be seen in Fig. S4 (ESI†). A brief description of the evolutions is given below. First, the MgH₂–AlH₃–CeF₃ composite is prepared by ball milling the starting materials of MgH₂, AlH₃ and CeF₃. After milling, the particle sizes of the starting materials decrease because both of the hydrides are brittle materials. Meanwhile, CeF₃ may cover the particle surfaces of MgH₂ and AlH₃. MgH₂ and AlH₃ particles also contact closely with each other during milling. Then, during the hydrogen desorption process (i.e. at the heating stage), AlH₃ first decomposes to form the oxide-free Al*, and this freshly formed Al* further reacts with MgH₂ to form the Mg solid solution (Mg_ss(Al)) and Mg₁₇Al₁₂, thus thermodynamically improving the desorption properties of MgH₂. On the other hand, CeF₃ also reacts with MgH₂ to form MgF₂ and CeH_2–3, which distribute surrounding the Mg–Al phase particles and act both as an impediment to the grain growth of Mg–Al phases and as hydrogen diffusion gateways. In this way, the MgF₂/CeH_2–3 layers kinetically improve the desorption properties of MgH₂. In a word, this schematic diagram shows a picture of how the thermodynamics and kinetics of MgH₂ are tailored by the co-addition of AlH₃ and CeF₃. The synergistic effect is believed to contribute to the reduction of desorption temperature, the absence of a desorption induction period, the acceleration of desorption kinetics and the extremely good cycling stability.

Finally, we carried out the thermal decompositions of the MgH₂ + mCeF₃ and MgH₂ + 0.1AlH₃ + nCeF₃ samples by DSC analysis (Fig. 6). It is found in Fig. 6a–c that the peak desorption temperatures of the MgH₂ + mCeF₃ samples generally locate at about 360 °C, even when the amount of added CeF₃ is increased. A study by Jin et al. also showed that the desorption temperature of MgH₂ doped with 1 mol% CeF₃ is only slightly reduced by less than 20 °C; this reduction is even less than that seen in the present work (about 30 °C). However, Fig. 6d–f show that in the presence of 0.1AlH₃, the peak desorption temperature of MgH₂ can be further reduced if the amount of added CeF₃ is increased. A reduction of about 70 °C is achieved for the MgH₂ + 0.1AlH₃ + 0.05CeF₃ sample compared to the pure MgH₂ (Fig. 1a).

Fig. 6 DSC curves of (a–c) MgH₂ + mCeF₃ (m = 0.01, 0.02, and 0.05) and (d–f) MgH₂ + 0.1AlH₃ + nCeF₃ (n = 0.01, 0.02, and 0.05) samples.

To preliminarily study the role of AlH₃ in the MgH₂ + 0.1AlH₃ + 0.05CeF₃ sample, we analysed the morphologies of the as-milled MgH₂ with or without AlH₃ addition. Fig. 7 shows the SEM images of the pure MgH₂ sample and the MgH₂ + 0.25AlH₃ sample after ball milling. It is seen that the particles of the as-milled MgH₂ + 0.25AlH₃ sample are generally smaller than those of the as-milled pure MgH₂ sample, which suggests that AlH₃ addition leads to the particle refinement of MgH₂. This may imply that more MgH₂ particle surfaces come into contact with the excess CeF₃. The increased contact means that more MgF₂/CeH_2–3 phases acting as hydrogen diffusion gateways are generated, thus further reducing the desorption temperature of MgH₂. It should be noted that this temperature reduction may originate from the kinetic enhancement rather than the destabilisation effect. Thus, hydrogen desorption analyses of the samples were carried out by DSC measurements with a relatively fast heating rate of 10 °C min⁻¹. Such a heating rate may lead to hysteresis in the DSC curves if the reaction kinetics are not fast enough. The Mg-based materials with added AlH₃ may possess more MgF₂/CeH_2–3 phases acting as hydrogen diffusion gateways, resulting in better kinetic properties. Therefore, the addition of AlH₃ in the present work is believed to have kinetically helped CeF₃ to affect the Mg-based materials. All in all, in addition to thermodynamically destabilising MgH₂, AlH₃ may also kinetically help CeF₃ to affect MgH₂. However, the exact role of AlH₃ should be carefully studied in detail by other analytic techniques.

Fig. 7 SEM images of the as-milled samples: (a) MgH₂ + 0.25AlH₃; and (b) MgH₂.

4 Conclusions

The hydrogen desorption properties including the kinetics, thermodynamics, and cycling stability of MgH₂ are significantly improved by the synergistic addition of AlH₃ and CeF₃. The formations of Mg solid solution and Mg₁₇Al₁₂ contribute to the thermodynamic destabilisation of MgH₂, while the formation of CeH_2–3 acting as a “hydrogen pump” and an impediment to grain growth of MgH₂ lead to the kinetic improvement and the extremely good cycling stability of MgH₂. Finally, it was found that AlH₃ kinetically helps CeF₃ to affect MgH₂. This work provides a method for simultaneously tailoring the thermodynamic and kinetic properties of MgH₂ by the synergistic addition of metal hydride and rare earth fluoride.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 51171168 and 51471149), Public Project of Zhejiang Province (no. 2013C31033), Key Science and Technology Innovation Team of Zhejiang province (no. 2010R50013), and Program for Innovative Research Team in University of Ministry of Education of China (IRT13037).

Notes and references

1. P. Selvam, B. Viswanathan, C. S. Swamy and V. Srinivasan, Int. J. Hydrogen Energy, 1986, 11, 169–192.
2. K. F. Aguey-Zinsou and J. R. Ares-Fernandez, Energy Environ. Sci., 2010, 3, 526–543.
3. F. Cheng, Z. Tao, J. Liang and J. Chen, Chem. Commun., 2012, 48, 7334–7343.
4. A. Zaluska, L. Zaluski and J. O. Strom-Olsen, J. Alloys Compd., 1999, 289, 197–206.
5. S. Bouaricha, J. P. Dodelet, D. Guay, J. Huot, S. Boily and R. Schulz, J. Alloys Compd., 2000, 297, 282–293.
6. G. Liang, J. Huot, S. Boily, A. Van Neste and R. Schulz, J. Alloys Compd., 1999, 292, 247–252.
7. C. X. Shang, M. Bououdina, Y. Song and Z. X. Guo, Int. J. Hydrogen Energy, 2004, 29, 73–80.
8. C. S. Zhou, Z. G. Z. Fang, J. Lu and X. Y. Zhang, J. Am. Chem. Soc., 2013, 135, 10982–10985.
9. P. Kalisvaart, E. Luber, H. Fritzsche and D. Mitlin, Chem. Commun., 2011, 47, 4294–4296.
10. J. Yuan, Y. Zhu and L. Li, Chem. Commun., 2014, 50, 6641–6644.
11. J. Cui, H. Wang, J. Liu, L. Ouyang, Q. Zhang, D. Sun, X. Yao and M. Zhu, J. Mater. Chem. A, 2013, 1, 5603–5611.
12. J. Cui, J. Liu, H. Wang, L. Ouyang, D. Sun, M. Zhu and X. Yao, J. Mater. Chem. A, 2014, 2, 9645–9655.
13. G. Xin, J. Yang, H. Fu, W. Li, J. Zheng and X. Li, RSC Adv., 2013, 3, 4167–4170.
14. Y. Liu, J. Zou, X. Zeng and W. Ding, RSC Adv., 2014, 4, 42764–42771.
15. T. Kelkar and S. Pal, J. Mater. Chem., 2009, 19, 4348–4355.
16. J. H. Dai, Y. Song and R. Yang, Int. J. Hydrogen Energy, 2011, 36, 12939–12949.
17. A. Zaluska, L. Zaluski and J. O. Strom-Olsen, Appl. Phys. A: Mater. Sci. Process., 2001, 72, 157–165.
18. H. Z. Liu, X. H. Wang, Y. A. Liu, Z. H. Dong, G. Z. Cao, S. Q. Li and M. Yan, J. Mater. Chem. A, 2013, 1, 12527–12535.
19. H. Z. Liu, X. H. Wang, Y. G. Liu, Z. H. Dong, H. W. Ge, S. Q. Li and M. Yan, J. Phys. Chem. C, 2014, 118, 37–45.
20. H. Z. Liu, X. H. Wang, Y. A. Liu, Z. H. Dong, S. Q. Li, H. W. Ge and M. Yan, J. Phys. Chem. C, 2014, 118, 18908–18916.
21. S. A. Jin, J. H. Shim, Y. W. Cho and K. W. Yi, J. Power Sources, 2007, 172, 859–862.
22. J. W. Kim, J. P. Ahn, S. A. Jin, S. H. Lee, H. S. Chung, J. H. Shim, Y. W. Cho and K. H. Oh, J. Power Sources, 2008, 178, 373–378.
23. A. Grzech, U. Lafont, P. C. Magusin and F. M. Mulder, J. Phys. Chem. C, 2012, 116, 26027–26035.
24. I. E. Malka, T. Czujko and J. Bystrzycki, Int. J. Hydrogen Energy, 2010, 35, 1706–1712.
25. I. E. Malka, M. Pisarek, T. Czujko and J. Bystrzycki, Int. J. Hydrogen Energy, 2011, 36, 12909–12917.
26. M. Ismail, N. Juahir and N. S. Mustafa, J. Phys. Chem. C, 2014, 118, 18878–18883.
27. M. Ismail, Int. J. Hydrogen Energy, 2014, 39, 2567–2574.
28. M. Ismail, Y. Zhao, X. B. Yu and S. X. Dou, RSC Adv., 2011, 1, 408–414.
29. M. F. Abdul Halim Yap, N. S. Mustafa and M. Ismail, RSC Adv., 2014 10.1039/C4RA12487A.
30. F. M. Brower, N. E. Matzek, P. F. Reigler, H. W. Rinn, C. B. Roberts, D. L. Schmidt, J. A. Snover and K. Terada, J. Am. Chem. Soc., 1976, 98, 2450–2453.
31. H. Z. Liu, X. H. Wang, Y. A. Liu and M. Yan, Chem. J. Chin. Univ., 2013, 34, 2274–2278.
32. H. Z. Liu, X. H. Wang, Z. H. Dong, G. Z. Cao, Y. A. Liu, L. X. Chen and M. Yan, Int. J. Hydrogen Energy, 2013, 38, 10851–10856.
33. L. Birks and H. Friedman, J. Appl. Phys., 1946, 17, 687–692.
34. Q. A. Zhang and H. Y. Wu, Mater. Chem. Phys., 2005, 94, 69–72.
35. J. C. Crivello, T. Nobuki, S. Kato, M. Abe and T. Kuji, J. Alloys Compd., 2007, 446, 157–161.
36. F. J. Liu and S. Suda, J. Alloys Compd., 1995, 231, 742–750.
37. C. X. Shang and Z. X. Guo, Int. J. Hydrogen Energy, 2007, 32, 2920–2925.
38. L. Z. Ouyang, X. S. Yang, M. Zhu, J. W. Liu, H. W. Dong, D. L. Sun, J. Zou and X. D. Yao, J. Phys. Chem. C, 2014, 118, 7808–7820.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1–4. See DOI: 10.1039/c4ra17139j

---