The hydrogen storage properties of MgH₂–Fe₇S₈ composites

Abstract

Nanostructured Fe₇S₈ was successfully synthesized and its catalytic effect on hydrogen absorption/desorption performance of MgH₂ is systemically discussed. The MgH₂ + 16.7 wt% Fe₇S₈ composite prepared by ball-milling method offers a striking catalytic activity for hydrogenation kinetics, and also reduces the initial decomposition temperature for MgH₂. The composite of MgH₂–Fe₇S₈ can absorb 4.000 wt% of hydrogen within 1800 s at 473 K, which is about twice that of pristine MgH₂ (1.847 wt%) under the same conditions. The onset hydrogen release temperature of Fe₇S₈-modified MgH₂ is 420 K, which is 290 K lower than that of additive-free MgH₂ (710 K). Meanwhile, the doped sample could release 4.403 wt% of hydrogen within 1800 s at 623 K as compared to 2.479 wt% of hydrogen by MgH₂. The activation energy for MgH₂–Fe₇S₈ is about 130.0 kJ mol⁻¹, approximately 36 kJ mol⁻¹ lower than that of MgH₂. The hydriding process of MgH₂ + 16.7 wt% Fe₇S₈ follows the nucleation and growth mechanism. The prominent hydrogen storage performances are related to the reactions between MgH₂ and Fe₇S₈. The newly formed MgS and Fe in the ball-milling process present a co-catalytic effect on the hydrogen storage performance of MgH₂.

---

1. Introduction

The global environmental issues related to the excessive emission of greenhouse gases and burning of fossil materials have driven us to center our attention on alternative energy sources that can be extensively developed and widely used.^1–4 Meanwhile, hydrogen energy is deemed as one of the most potential and ideal energy alternatives on account of its high energy density and clean and environmentally friendly nature.^5–8 However, the practical application and popularization of hydrogen is still hindered by numerous complicated technical hurdles. In recent years, solid-state storage materials have been receiving worldwide attention as ideal hydrogen storage media. Among them, magnesium hydride (MgH₂), which has a high hydrogen storage capacity, excellent reversibility and abundant availability, is regarded as one of the most marvelous candidates for potential hydrogen storage.^9–12 Unfortunately, the widespread practical application of MgH₂ is retarded by sluggish dynamic performance and high thermodynamic operating temperature.

Various attempts covering mechanical alloying,^13,14 nano-structuring,¹⁵ and catalyst doping^16–20 have been made by numerous groups to enhance the thermodynamic and hydrogen absorption and desorption properties. Among the above-mentioned modified systems, doping MgH₂ with catalysts has been considered as an efficient way to boost the properties of MgH₂.^21–23 Especially, the transition metal iron (Fe) and its compounds have attracted considerable interest due to their pronounced catalytic effect on MgH₂. Yan et al.¹⁵ observed a lower onset dehydrogenation temperature for the MgH₂–10 wt% FeB/CNTs composite; the modified composites could start to release hydrogen at about 196 °C and absorb 6.2 wt% of H₂ at 150 °C within 10 min. Chen et al.²⁴ found that doping MgH₂ with 5 wt% Fe nanosheets reduced the onset desorption temperature, and hydrogen could be released at 182.1 °C and absorbed at 75 °C by the composite. Gattia et al.²⁵ reported that Fe and its oxides are suitable catalysts for hydrogen storage as they drastically speed up the reaction kinetics.

In addition to iron and its compounds, sulfide compounds also showed remarkable catalytic activities for the improvement of sorption kinetics and thermodynamic properties of MgH₂. Jia et al.²⁶ discovered that MgH₂ with 16.7 wt% of MoS₂ additive showed outstanding hydrogenation/dehydrogenation properties and the hydrogenation activation energy of the modified composites was determined to be 87.19 ± 4.48 kJ mol⁻¹, lower than that of the pure MgH₂ (111.03 ± 3.41 kJ mol⁻¹). WS₂ was also found to promote the hydrogenation/dehydrogenation behavior of MgH₂ by Wang et al.²⁷ Our previous work reported that the hydrogen intake/uptake performance of MgH₂ was greatly influenced by nano-structured polysulfides containing iron (Fe₃S₄ and FeS₂).^28,29 The polysulfide containing iron ion is a kind of compound in polyvalent states, which can be approximately considered by the compound of ferrous sulfide and iron sulfide. Multi-valent catalysts also showed great potential in enhancing de-/hydrogenation kinetics of MgH₂ as reported by Xie and co-workers.³⁰ Meanwhile, the excellent hydrogen storage behavior of MgH₂ might be related to the unstable Fe²⁺. Fe₇S₈ is also a polysulfide containing iron; multi-valent and unstable Fe²⁺ exist in Fe₇S₈, which might contribute to the enhancement of the hydrogen storage capacity of MgH₂.

Given the encouraging results mentioned above, in this work, Fe₇S₈ was successfully synthesized and its catalytic effectiveness toward hydrogen adsorption/desorption of MgH₂ was systemically investigated. Besides, the corresponding catalytic mechanism of Fe₇S₈ to the enhancement of hydrogen properties of the pristine MgH₂ was also presented based on the information reflected by hydrogen absorption/desorption kinetics, phase compositions and microstructure observations.

2. Experimental

2.1 Synthesis of Fe₇S₈

1 mmol FeSO₄·7H₂O and 2 mmol thiourea were added to a solvent composed of triethylenetetramine and deionized water (the ratio of the components of the solvent was 1 : 2) and stirred for 1 h by magnetic stirring until the solution was homogenized. The solution was then transferred into a Teflon autoclave and maintained at 140 °C for 24 h. After cooling down to room temperature naturally, the precipitate was gathered by centrifugation and washed with distilled water and absolute alcohol several times. The obtained precipitate was dried in a vacuum oven at 80 °C for 6 h. The precursor [Fe₁₈S₂₅](TETA)₁₄ was finally prepared.

0.1 g precursor [Fe₁₈S₂₅](TETA)₁₄ and 15 ml oleylamine were added into a boiling flask and promptly heated to 90 °C in a vacuum atmosphere. To protect the sample from oxidation, argon was introduced into the boiling flask as a protective gas after the vacuum was stopped. The sample was then quickly heated to 260 °C and kept warm for 8 h. After the reaction was complete and the temperature of the boiling flask decreased to room temperature, extra 40 ml alcohol was introduced into the flask under mechanical stirring for about 10 min and the solution was further aged for 2 h. Fe₇S₈ was successfully synthesized and could be separated after the aged solution was centrifuged and washed with absolute alcohol.

2.2 Synthesis of MgH₂–Fe₇S₈ composites

MgH₂ was synthesized by the method of hydrogen combustion from commercial magnesium powder. Magnesium was hydrogenated at 400 °C with a hydrogen pressure of 4 MPa for 10 h. The obtained powder was then ball-milled at 350 rpm for 2 h. The target MgH₂ was successfully prepared after hydrogenating the sample one more time at 400 °C under 4 MPa hydrogen pressure for 10 h.

MgH₂–Fe₇S₈ composite was synthesized through the ball-milling method by mechanical milling of MgH₂ and Fe₇S₈ with a weight ratio of 5 : 1. To reduce the increasing temperature, the milling process was launched at a speed of 350 rpm and ball-milled in one direction for 0.5 h and paused for about 0.25 h, unceasingly starting in another direction for 0.5 h. The ball-to-sample weight ratio was 20 : 1. To prevent the sample from oxygen and vapor, all samples were handled in an air-filled glove box.

2.3 Characterization

X-ray diffraction (XRD) was performed on a SmartLab high-resolution X-ray diffractometer (made by Rigaku Company) with Cu Kα radiation at 40 kV, 40 mA. The scanning speed was 4° min⁻¹ in the range of 10° to 80°. Scanning electron microscopy (SEM) was employed to observe the microstructure and morphology of the formed sample. The hydrogen absorption/desorption behavior in relation to temperature for the composite of MgH₂–Fe₇S₈ and non-additive MgH₂ was determined in a pressure–composition–temperature (PCT) apparatus. The sample was gradually heated from room temperature to the set temperature, which was kept constant during the whole experiment. Temperature–programmed–desorption (TPD) was used to investigate the initial desorption temperature for the composite of MgH₂–Fe₇S₈ and MgH₂ free-additive. All the tested composites were heated from room temperature to 770 K at a heating rate of 10 K min⁻¹. Differential scanning calorimetry (DSC) was performed from 200 K to 770 K at various heating rates: 5 K min⁻¹, 10 K min⁻¹, 15 K min⁻¹, and 20 K min⁻¹.

3. Results and discussion

3.1 Characteristics of the prepared nano-structured Fe₇S₈

XRD pattern of the as-synthesized Fe₇S₈ in Fig. 1a exhibits dominant diffraction peaks at 2θ = 29.9°, 33.8°, 43.9°, 53.1°, 57.3°, 64.8°, and 71.2° matching well with the Fe₇S₈ phase, in accordance with the data reported in JCPDS PDF 33-0664. As can be seen from Fig. 1b, the prepared sample presents the morphology of two-dimensional hexagonal sheets with a thickness of approximately 20 nm and a diameter of around 250 nm. Based on XRD and SEM results, nano-structured Fe₇S₈ was successfully fabricated.

Fig. 1 (a) XRD pattern of the as-synthesized Fe₇S₈ (b) SEM image of the as-synthesized Fe₇S₈.

3.2 Catalytic effectiveness of the fabricated Fe₇S₈ for the dehydrogenation/hydrogenation kinetics of MgH₂

The catalytic role of the as-synthesized Fe₇S₈ in the dehydrogenation/hydrogenation kinetics of MgH₂ was studied by using TPD isothermal dehydrogenation/hydrogenation measurements. The TPD curves of MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ composite at a heating rate of 10 K min⁻¹ are shown in Fig. 2. As observed, the Fe₇S₈ doped MgH₂ sample shows a striking effect of reducing the onset desorption temperature of MgH₂. The prepared MgH₂ commences to release hydrogen at approximately 710 K, whereas the onset desorption temperature for Fe₇S₈-modified MgH₂ presents a remarkable decrease to 420 K, which is 290 K lower than that of additive-free MgH₂. As for the pristine MgH₂, the full desorption capacity is about 4.45 wt%. After introducing nano-Fe₇S₈ into MgH₂, the desorbed content is increased and the maximum yield for the desorption process is nearly 5.72 wt%. Based on TPD measurements, the lower initial desorption temperature and the maximum desorption yield reveal that the introduced nano-Fe₇S₈ into MgH₂ could successfully boost the desorption properties of MgH₂.

Fig. 2 Thermal decomposition curves of MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ composite.

Isothermal dehydrogenation experiments were conducted at 623 K to further compare the effect of the doped Fe₇S₈ on the desorption kinetics process in MgH₂; the relative curves for MgH₂ with and without additives in the desorption process are presented in Fig. 3. It is evident that with the increase in temperature, the hydrogen desorption properties of MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ are enhanced. MgH₂ + 16.7 wt% Fe₇S₈ composite exhibits much faster kinetics than that of pristine MgH₂. The MgH₂ + 16.7 wt% Fe₇S₈ sample could release 1.515 wt% H₂ in 1800 s at 573 K. In contrast, the pure MgH₂ only liberated 0.250 wt% H₂ in the same time period. When the temperature increases to 623 K, the additive-free MgH₂, which could only desorb 2.479 wt% of hydrogen in 1800 s at 623 K, shows worse desorption properties than the doped sample. MgH₂ + 16.7 wt% Fe₇S₈ composite could liberate 4.403 wt% of hydrogen within 1800 s under identical conditions. As compared to non-additive MgH₂, the dopant Fe₇S₈ promotes the dehydrogenation kinetics.

Fig. 3 Dehydrogenation properties of MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ composite at 573 K, 623 K.

The Kissinger's method³¹ was performed to further understand the enhanced dehydrogenation process for MgH₂ and MgH₂–Fe₇S₈ composites; the apparent activation energy (E_a) of those two composites were estimated by the Arrhenius equation described as below:

where α is the heating rate (K min⁻¹), T_m is the peak temperature for the maximum desorption rate (K), and R is the gas constant of 8.314 J (mol⁻¹ K⁻¹). Fig. 4a and b display DSC curves of MgH₂ and MgH₂–Fe₇S₈ composites, respectively, at various heating rates. From the slope of the fitted line in Fig. 4c, the slope for the two samples were obtained as −19.95 and −15.64, while the corresponding apparent activation energy E_a for MgH₂ was approximately 165.9 kJ mol⁻¹. It can be noticed that the apparent activation energy E_a of MgH₂–Fe₇S₈ was decreased to 130.0 kJ mol⁻¹, which is lower by about 36 kJ mol⁻¹ than that of MgH₂. The reduced apparent activation energy demonstrated above gives the indication that the dopant Fe₇S₈ contributes to decreasing the energy barrier during the desorption process of MgH₂, which is directly responsible for the superior improvement of the dehydrogenation properties. In a recent literature,²⁵ activation energies for milled MgH₂ + 5 wt% Fe, MgH₂ + 5 wt% Fe₂O₃ and MgH₂ + 5 wt% Fe₃O₄ calculated using the Kissinger plot were 220.69 kJ mol⁻¹, 231.90 kJ mol⁻¹ and 304.45 kJ mol⁻¹, respectively, which indicates that the addition of Fe₇S₈ may show superior catalytic effect compared to that of some other additives.

Fig. 4 DSC curves of (a) pure MgH₂ and (b) MgH₂ + 16.7 wt% Fe₇S₈ composites with a heating rate of 5 K min⁻¹, 10 K min⁻¹, 15 K min⁻¹. (c) The corresponding Kissinger plots for the hydrogen desorption reaction for MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ composites at various heating rates.

In addition to the hydrogen desorption behavior, the modified impact of Fe₇S₈ on the hydrogen uptake performance on MgH₂ was also investigated by reabsorbing the dehydrogenated MgH₂ and MgH₂–Fe₇S₈ composites under 3 MPa of H₂. Fig. 5 presents the hydrogenation curves of MgH₂ and MgH₂–Fe₇S₈ composite at 473 K and 523 K, respectively. It can be clearly seen that the dehydrogenated MgH₂ sample only absorbs 1.847 wt% of hydrogen within 1800 s at 473 K, whereas the dehydrogenated MgH₂ catalyzed by Fe₇S₈ composite shows higher hydrogen absorption ability under the same conditions, with an absorption capacity of 4.000 wt%. When the temperature sharply increases to 523 K, the hydrogen absorption ability of both dehydrogenated MgH₂–Fe₇S₈ composite and additive-free MgH₂ presents an obvious rise. The amount of hydrogen uptake increases to 3.915 wt% for pristine MgH₂ within 1800 s, while that of the Fe₇S₈ doped composite reaches 4.804 wt%.

Fig. 5 Hydrogenation properties of (a) MgH₂ at 473 K; (b) MgH₂ + 16.7 wt% Fe₇S₈ composite at 473 K; (c) MgH₂ at 523 K; (d) MgH₂ + 16.7 wt% Fe₇S₈ composite at 523 K.

The hydrogenation mechanism can usually be explained by comparing the hydrogen absorption rate curves with the rate equations for MgH₂ + 16.7 wt% Fe₇S₈ and MgH₂ composites. The Avrami–Erofeev equation, eqn (2), is usually employed to fit the hydrogenation absorption process, which gives a strong idea for the nucleation and growth processes.

where α is the reacted fraction, k is the rate constant, and m is the order of the reaction. Fig. 6 shows the fitted hydrogenation kinetic curves of MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ composites at 523 K. It is obvious that there is a great difference between the fitted curve and the normalized curve for MgH₂ without additives. However, this similar phenomenon does not appear in the composite of MgH₂–Fe₇S₈, which is consistent with the experimental data. The fitted curves in Fig. 6 demonstrate that the reaction mechanism of MgH₂–Fe₇S₈ composite conforms to nucleation and growth processes. Besides, m represents the order of reaction; the rate-controlling steps for the hydrogen uptake process are connected with the value of m. It has been reported that the values of m for the one-dimensional diffusion process and the three-dimensional interfacial reaction are assigned as 0.620 and 1.070, respectively.³² As for MgH₂ without the catalyst, m is 0.9265 at 523 K, which is pretty close to 1.070. For the doped MgH₂–Fe₇S₈ composite, the value of m was 0.4212, which approaches 0.620. Therefore, the rate-controlling steps of MgH₂ during the hydrogenation process changes from the three-dimensional interfacial reaction to the one-dimensional diffusion process, which is relevant to the addition of Fe₇S₈.

Fig. 6 Fitted hydrogenation kinetic curves of MgH₂ and MgH₂ + 16.7 wt% Fe₇S₈ composites at 523 K under the pressure of 3 MPa.

3.3 Structural characterization and catalytic effect mechanism of MgH₂–Fe₇S₈ composite

Fig. 7 presents SEM images of MgH₂ and MgH₂–Fe₇S₈ composite after the re-hydrogenation process. It is obvious that the particle size of pristine MgH₂ is much bigger than that of MgH₂–Fe₇S₈ composite; some of the particles in pristine MgH₂ are agglomerated and inhomogeneous. The particles in the Fe₇S₈ added MgH₂ have a loose structure and are highly dispersed, while the particles of additive-free MgH₂ aggregate into large clusters and become tough. Xie et al.³³ and Varin et al.³⁴ reported that the decreased particle size and loose structure could boost the rates of dehydrogenation-hydrogenation, which indicates that the added Fe₇S₈ can not only help to decrease the particle size but also serve as an inhibitor to limit the formation of large clusters and result in favourable hydrogen kinetics.

Fig. 7 SEM images of (a) MgH₂, (b) MgH₂–Fe₇S₈ composite after the re-hydrogenation process.

To further elaborate on the catalytic mechanism related to the striking hydrogen absorption/desorption kinetics and thermodynamic properties of MgH₂–Fe₇S₈ composite, XRD analysis was employed to characterize the phase structure of the doped composite at different stages. Fig. 8 shows the XRD patterns of MgH₂–Fe₇S₈ composites that were collected at the ball-milling and de/hydrogenation stages. In the XRD pattern of the composite at the ball-milling stage, the main diffraction peak assigned to MgH₂ appears accompanied by some new phases of Mg₂FeH₆, MgS, and Fe. In the XRD pattern of the composite at the hydrogenation stage, all the diffraction peaks are very intense due to hydrogenation. After the re-hydrogenation procedure, the major diffraction peaks assigned to Fe, Mg₂FeH₆ and MgS are still present. The appearance of new phases of Mg₂FeH₆, MgS and Fe indicates the reaction of MgH₂ and Fe₇S₈ during the ball-milling/re-hydrogenation stage. The reactions can be summarized as follows:

8MgH₂ + Fe₇S₈ → 7Fe + MgS + 8H₂ (1)

Fe + 2MgH₂ + H₂ → Mg₂FeH₆ (2)

Fig. 8 XRD patterns of MgH₂–Fe₇S₈ composites at the (a) ball-milling (b) hydrogenation (c) dehydrogenation stages.

To confirm the possibility of the reaction between MgH₂ and Fe₇S₈, the total change ΔG^θ was calculated. The standard Δ_fH^θ_m for MgH₂, Fe₇S₈, and MgS is −75.3, −824.2, and −346 kJ mol⁻¹, and Δ_fG^θ_m are −35.9, −742.2, and −341.8 kJ mol⁻¹, respectively. The total change ΔG^θ for reaction (1) is −1705.0 kJ mol⁻¹, whereas ΔG^θ for MgH₂ is −35.9 kJ mol⁻¹; this confirms the possibility of occurrence of reaction (1) from the thermodynamic potentials, and the dopant Fe₇S₈ is conducive to the hydrogen absorption/desorption process.

In the XRD pattern of the composite at the dehydrogenation stage, the dominant diffraction peak belongs to Mg, implying that Mg₂FeH₆ and MgH₂ produced in the hydrogenation process have completely decomposed. Some weak peaks corresponding to MgO are found during the whole stage, which is ascribed to the oxidation of Mg in the test operation process. The existing equation in the hydrogen release stage is as follows:

MgH₂ → Mg + H₂ (3)

Mg₂FeH₆ → Fe + 2Mg + 3H₂ (4)

The formation of the intermediate phase of Mg₂FeH₆ has shown a striking catalytic role in the hydrogen sorption behavior. Xiao et al.³⁵ reported that the Mg₂FeH₆@MgH₂ core–shell sample exhibited faster hydrogen desorption kinetics, releasing more than 5.0 wt% H₂ within 50 min at 280 °C. Apart from the Mg₂FeH₆–MgH₂ system, Mg₂FeH₆ combined with NaBH₄³⁶ composite has boosted dehydrogenation properties, while the dehydriding temperature of NaBH₄ was reduced by at least 150 K when combined with Mg₂FeH₆. The enhancement of the dehydriding properties of LiBH₄ was also confirmed by Li et al.³⁷ Mg₂FeH₆ is a hydrogen-storage material, which has an excellent ability for hydrogen uptake/release performance.³⁸ As an intermediate, Mg₂FeH₆ can uptake and liberate hydrogen. Therefore, the produced Mg₂FeH₆ plays a critical role in the improvement of the hydrogen storage properties.

The schematic diagram for the catalytic mechanism of Fe₇S₈ during the hydrogenation/dehydrogenation processes on MgH₂ is shown in Fig. 9. During the hydrogenation process, the phases of Fe, MgH₂ and H₂ are generated by the formation of the intermediate phase of Mg₂FeH₆. In the dehydrogenation stage, Mg₂FeH₆ produces Mg and Fe. It is suggested that Fe as an active species exists in the whole dehydrogenation/hydrogenation process together with the newly formed MgS. Apart from that, from our previous investigation, the hydrogen absorption/desorption performance of the MgH₂ + Fe and MgH₂ + MgS systems is inferior to that of the in situ formed system.²⁹ Therefore, the newly produced Fe and MgS during the ball-milling stage may co-catalyze the hydrogen storage performance of the MgH₂–Fe₇S₈ system.

Fig. 9 Schematic diagram of the catalytic mechanism of Fe₇S₈ during the hydrogenation/dehydrogenation processes on MgH₂.

4. Conclusions

Nano-Fe₇S₈ was successfully fabricated and its catalytic effect on the hydrogen storage performance of MgH₂ was systemically investigated. The isothermal hydrogen absorption/desorption measurement shows that the dehydrogenated MgH₂ catalyzed by Fe₇S₈ composite has the hydrogen absorption capacity of 4.000 wt% within 1800 s at 473 K; only 1.847 wt% hydrogen could be absorbed by MgH₂ under the same conditions. During the dehydrogenation process, MgH₂ + 16.7 wt% Fe₇S₈ composite could release 4.403 wt% of hydrogen within 1800 s at 623 K as compared to 2.479 wt% of hydrogen by MgH₂. The onset desorption temperature for Fe₇S₈-modified MgH₂ is 420 K, lower than that of additive-free MgH₂ (710 K). The hydriding process of MgH₂ + 16.7 wt% Fe₇S₈ follows the nucleation and growth mechanism. The activation energy E_a of MgH₂–Fe₇S₈ is about 130.0 kJ mol⁻¹, approximately 36 kJ mol⁻¹ lower than that of MgH₂. The striking catalytic effect of Fe₇S₈ on the hydrogen storage performance of MgH₂ is related to the reactions between MgH₂ and Fe₇S₈. The newly formed MgS and Fe during the ball-milling process present a co-catalytic effect on the hydrogen storage performance of MgH₂.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the Natural Science Foundation of Hebei Province of China (No. E2019415036); Science and Technology Project of Hebei Education Department (BJ2020043); Science and Technology Support Plan of Qinhuangdao Science and Technology Bureau (201805A010); Doctoral Foundation of Hebei University of Environmental Engineering (201805)

Notes and references

1. Z. W. Ma, J. X. Zou, D. Khan, W. Zhu, C. Z. Hu, X. Q. Zeng and W. J. Ding, J. Mater. Sci. Technol., 2019, 35, 2132–2143.
2. M. Liu, S. Zhao, X. Xiao, M. Chen, C. Sun, Z. Yao, Z. Hu and L. Chen, Nano Energy, 2019, 61, 540–549.
3. X. Xie, C. Ni, B. Wang, Y. Zhang, X. Zhao, L. Liu, B. Wang and W. Du, J. Alloys Compd., 2020, 816, 152634.
4. Z. Shao, Y. Li, C. Liu, W. Ai, S.-P. Luo and Q. Liu, Nat. Commun., 2020, 11, 591.
5. J. Du, Z. Lan, H. Zhang, S. Lu, H. Liu and J. Guo, J. Alloys Compd., 2019, 802, 660–667.
6. X. Yao, C. Wu, A. Du, J. Zou, Z. Zhu, P. Wang, H. Cheng, S. Smith and G. Lu, J. Am. Chem. Soc., 2007, 129, 15650–15654.
7. L. Dan, L. Hu, H. Wang and M. Zhu, Int. J. Hydrogen Energy, 2019, 44, 29249–29254.
8. Z. Cao, L. Ouyang, H. Wang, J. Liu, L. Sun, M. Felderhoff and M. Zhu, Int. J. Hydrogen Energy, 2016, 41, 11242–11253.
9. M. Liu, X. Xiao, S. Zhao, S. Saremi-Yarahmadi, M. Chen, J. Zheng, S. Li and L. Chen, Int. J. Hydrogen Energy, 2019, 44, 1059–1069.
10. Y. Zheng, K. Li, H. Wang, D. Tian, Y. Wang, X. Zhu, Y. Wei, M. Zheng and Y. Luo, Appl. Catal., B, 2017, 202, 51–63.
11. H.-J. Lin, J. Matsuda, H.-W. Li, M. Zhu and E. Akiba, J. Alloys Compd., 2015, 645, S392–S396.
12. G. Gizer, J. Puszkiel, M. V. C. Riglos, C. Pistidda, J. M. Ramallo-Lopez, M. Mizrahi, A. Santoru, T. Gemming, J. C. Tseng, T. Klassen and M. Dornheim, Sci. Rep., 2020, 10, 12.
13. G. Liang, J. Huot, S. Boily, A. V. Neste and R. Schulz, J. Alloys Compd., 1999, 292, 247–252.
14. M. Y. Song, S. N. Kwon, S. H. Hong, C. G. Park, S. H. Baek, H. S. Kim, D. R. Mumm and J. S. Bae, Catal. Today, 2007, 120, 281–286.
15. S. Gao, X. Wang, H. Liu, T. He, Y. Wang, S. Li and M. Yan, J. Power Sources, 2019, 438, 227006.
16. J. Zhang, L. He, Y. Yao, X. J. Zhou, L. P. Yu, X. Z. Lu and D. W. Zhou, Renewable Energy, 2020, 154, 1229–1239.
17. F. A. H. Yap, N. N. Sulaiman and M. Ismail, Int. J. Hydrogen Energy, 2019, 44, 30583–30590.
18. N. H. Idris, N. S. Mustafa and M. Ismail, Int. J. Hydrogen Energy, 2017, 42, 21114–21120.
19. N. A. Ali, N. H. Idris, M. F. M. Din, N. S. Mustafa, N. A. Sazelee, F. A. H. Yap, N. N. Sulaiman, M. S. Yahya and M. Ismail, RSC Adv., 2018, 8, 15667–15674.
20. N. A. Sazelee, N. H. Idris, M. F. Md Din, M. S. Yahya, N. A. Ali and M. Ismail, Results Phys., 2020, 16, 102844.
21. M. Ismail, M. S. Yahya, N. A. Sazelee, N. A. Ali, F. A. H. Yap and N. S. Mustafa, J. Magnes. Alloys, 2020, 8, 832–840.
22. T. B. Andres, M. Z. Luis and M. Marcos, Int. J. Hydrogen Energy, 2020, 45, 27421–27433.
23. N. A. Ali, N. H. Idris, M. F. M. Din, M. S. Yahya and M. Ismail, J. Alloys Compd., 2019, 796, 279–286.
24. L. Zhang, L. Ji, Z. Yao, N. Yan, Z. Sun, X. Yang, X. Zhu, S. Hu and L. Chen, Int. J. Hydrogen Energy, 2019, 44, 21955–21964.
25. D. M. Gattia, M. Jangir and I. P. Jain, J. Alloys Compd., 2019, 801, 188–191.
26. Y. Jia, S. Han, W. Zhang, X. Zhao, P. Sun, Y. Liu, H. Shi and J. Wang, Int. J. Hydrogen Energy, 2013, 38, 2352–2356.
27. J. S. Wang, W. Zhang, Y. Cheng, D. D. Ke and S. M. Han, J. Wuhan Univ. Technol., Mater. Sci. Ed., 2015, 30, 670–673.
28. W. Zhang, Y. Cheng, D. Han and S. Han, Energy, 2015, 93(pt 1), 625–630.
29. W. Zhang, G. Xu, Y. Cheng, L. J. Chen, Q. Huo and S. Y. Liu, Dalton Trans., 2018, 47, 5217–5225.
30. L. S. Xie, J. S. Li, T. B. Zhang and L. Song, Mater. Charact., 2017, 133, 94–101.
31. H. E. Kissinger, Anal. Chem., 1957, 29, 1702–1706.
32. J. S. Pedersen, J. Appl. Crystallogr., 1994, 27, 595–608.
33. L. Xie, Y. Liu, Y. T. Wang, J. Zheng and X. G. Li, Acta Mater., 2007, 55, 4585–4591.
34. R. A. Varin, T. Czujko, E. B. Wasmund and Z. S. Wronski, J. Alloys Compd., 2007, 432, 217–231.
35. X. Xiao, C. Xu, J. Shao, L. Zhang, T. Qin, S. Li, H. Ge, Q. Wang and L. Chen, J. Mater. Chem. A, 2015, 3, 5517–5524.
36. G. Q. Li, M. Matsuo, S. Deledda, B. C. Hauback and S. Orimo, Mater. Trans., 2014, 55, 1141–1143.
37. G. Li, M. Matsuo, S. Deledda, R. Sato, B. Hauback and S.-I. Orimo, Mater. Trans., 2013, 54, 1532–1534.
38. Y. Wang, F. Cheng, C. Li, Z. Tao and J. Chen, J. Alloys Compd., 2010, 508, 554–558.

---

Footnote

† Both authors contributed equally to this work.

---