Graphene decorated with Fe nanoclusters for improving the hydrogen sorption kinetics of MgH₂ – experimental and theoretical evidence

Abstract

Graphene decorated with Fe clusters is proposed to be a possible alternative catalyst for the hydrogenation and dehydrogenation reactions of MgH₂. In particular, graphene decorated with Fe clusters is effective for both hydrogenation and dehydrogenation processes of MgH₂. The change in enthalpy and entropy values of hydrogen absorption determined for MgH₂ with 5 wt% graphene decorated with Fe clusters is −50.4 ± 2.9 kJ per mole of H₂ and 99.8 ± 5.2 J K⁻¹ per mole of H₂, respectively. This is significantly lower than those for well-established metal catalysts and nano-interfacial confined MgH₂. Moreover, the graphene decorated with Fe clusters facilitates the fast rehydrogenation kinetics of MgH₂, which reabsorbed 90% of the total reabsorption capacity in less than 4 minutes at 300 °C and 20 atm. In addition, TEM analysis reveals that MgH₂ particles are covered by graphene with Fe clusters, resulting in the reduction of grain growth. Density functional theory shows that the defects in graphene act as the active sites for the dehydrogenation of MgH₂, while the Fe clusters reduce the adsorption of dissociated H atoms, resulting in low-temperature dehydrogenation. Thus, graphene decorated with metal clusters could open up a new way of designing a new type of catalyst which could replace transition metal catalysts.

---

Introduction

Functionalized graphene has quickly caught much attention due to its unexpected properties in the field of catalysis.¹ Graphene oxide is a successful modification of graphene, where graphene oxide shows high catalytic activity during oxidation processes.^2,3 In the same fashion, metal cluster-decorated graphene happens to have unexpected catalytic effects on various systems.^4–6 In particular, graphene decorated with iron clusters is experimentally and theoretically found to store hydrogen.^7,8 During the hydrogen absorption process over graphene with iron clusters, calculations also revealed that H₂ dissociation occurs over iron clusters, which is considered to cause spillover towards graphene.⁷ Such a phenomenon can be a key reaction for the hydrogenation process as H₂ dissociation is generally the first reaction step of the hydrogenation reaction. Here, magnesium hydride (MgH₂) is chosen to evaluate the catalytic effect of graphene with Fe nanoclusters. Magnesium hydride is a potential hydrogen storage material containing 7.6 wt% H₂ (theoretical capacity). However, an effective catalyst is required to overcome the high thermodynamic stability and slow hydrogenation reaction kinetics of MgH₂. The reaction of magnesium hydride is a prototype hydrogenation reaction which requires Mg, H₂ and MgH₂ dissociation. In order to improve the kinetics of the hydrogenation and dehydrogenation of magnesium hydride, tremendous amounts of catalysts have been extensively investigated.^9–14 However, the search for better catalysts to accelerate the kinetics, hydrogenation, and dehydrogenation of MgH₂ is still a problematic issue for further practical applications. Metal-decorated graphene catalysts could be proposed to be potential catalysts for the hydrogenation reaction of such systems. The catalytic effect of graphene decorated with iron clusters is experimentally and theoretically investigated.

In particular, MgH₂ is used as a prototype reaction in order to evaluate the catalytic effect of graphene decorated with iron towards the hydrogenation and dehydrogenation process. Hydrogen absorption and desorption of MgH₂ with graphene decorated with iron clusters are experimentally tested. Transmission electron microscope imaging is used to observe the microstructure of graphene, iron, and MgH₂. Density functional theory is implemented in order to reveal the physical and chemical origin of the reaction mechanism. Although the study utilizes iron-decorated graphene and MgH₂, the catalytic effect and mechanism of graphene with metal systems should be generalized to further develop metal-decorated graphene catalysts.

Experimental

MgH₂ powder of 98% purity and Fe nanopowder (APS 10–30 nm) of 99.9% purity were purchased from Alfa Aesar. Graphene decorated with Fe nanoclusters (G-Fe) was synthesized in an electric arcing chamber using graphite electrodes. To ablate the material, an electric arc was generated using a DC source of 100 A and 25 V between a pure graphite rod (cathode) placed opposite to a grounded anode rod made of graphite impregnated with Fe nanopowder in an arcing chamber filled with Ar (99.999% purity) with a partial pressure of ~300 torr. After 8–10 min of the arcing process, a flake-like carbon material was found to be deposited around the walls of the arcing chamber which was later confirmed as G-Fe. The MgH₂ powder was ball-milled together with 5 wt% G-Fe under 5 atm hydrogen pressure in a custom fabricated stainless steel vial of 250 cm³ volume (capable of retaining up to 60 atm pressure) using a Retsch PM 400 planetary ball-miller at an operating speed of 150 rpm for a period of 60 minutes (6 × 10 minutes ball-milling with 15 minute interval in-between). The ball-to-powder ratio was kept at 30 : 1, where the weight of the sample was around 2.5 grams. Ball-milling under hydrogen atmosphere prevents MgO formation.¹³ For comparative analysis, MgH₂ (without G-Fe) was ball-milled under identical experimental conditions for 5 hours (30 × 10 minutes ball-milling with 15 minute interval in-between). Handling of the samples was done in an mBraun MB10 compact Ar-filled glove box (H₂O and O₂ levels <1 ppm).

The dehydrogenation and reabsorption kinetics of the samples were analyzed through temperature-programmed desorption (TPD) and pressure–composition isotherm (PCI) measurement using an automated four channel volumetric Sievert-type apparatus supplied by Advanced Materials Corporation. TPD analysis of the sample was done at an initial pressure of 10⁻³ torr under dynamic heating conditions with an accuracy of ±0.2 °C. The rehydrogenation kinetics was measured at different temperatures by charging the dehydrogenated MgH₂ samples with 20 atm hydrogen pressure. Thermal analyses of the samples were conducted using a differential scanning calorimeter (DSC 8000, PerkinElmer). The samples were heated from room temperature to about 500 °C with a set heating rate under flowing argon of 20 ml min⁻¹.

Structural characterization of the samples was performed by X-ray diffraction using a PANalytical X'Pert PRO diffractometer with a Cu K_α beam (λ = 1.5415 Å) operated at 40 kV and 40 mA. The samples were loaded in airtight sample holders sealed by a fine layer of parafilm (Pechiney plastic packing) to prevent the sample from being contaminated by oxygen and moisture.¹⁴ The microstructures of the samples were analyzed through bright field imaging and selected-area electron diffraction (SAED) using a FEI Tecnai 20G² transmission electron microscope (TEM) operated at 200 keV.

Results and discussion

TPD of MgH₂ samples was performed from room temperature (25 °C) to 450 °C under dynamic heating conditions at a heating rate of 2 °C min⁻¹. A comparison of the TPD profiles of pristine MgH₂, 5 hour ball-milled MgH₂ (un-catalyzed) and MgH₂ + 5 wt% G-Fe is shown in Fig. 1(a) & (b).

Fig. 1 Comparative TPD profiles of MgH₂ (a) hydrogen release capacity and (b) peak temperature.

It has been observed from the TPD profiles that the dehydrogenation temperature of MgH₂ + 5 wt% G-Fe is significantly lower than those of pristine and ball-milled MgH₂. The peak temperature of MgH₂ + 5 wt% G-Fe is 281.7 °C, whereas that of 5 hour ball-milled MgH₂ is 322.3 °C. This suggests that graphene together with Fe nanoclusters exhibits superior catalytic effect in improving the dehydrogenation behaviour of MgH₂.

Fig. 2(a) & (b) present the DSC curves of 5 hour ball-milled (un-catalyzed) MgH₂ and MgH₂ + 5 wt% G-Fe, respectively, determined at different heating rates (2, 5, 10 and 15 °C min⁻¹). The apparent desorption activation energy (E_a) of ball-milled MgH₂ and MgH₂ + 5 wt% G-Fe was determined using the Kissinger method.¹⁵

Fig. 2 DSC curves measured at different heating rates: (a) ball-milled MgH₂ and (b) MgH₂ + 5 wt% G-Fe; (c) ln(β/T²_P) vs. 1/T_P plot.

The peak desorption temperature (TPD) determined from the DSC curves and the corresponding heating rates (β) are the two parameters required for determining E_a using the Kissinger equation given below.

ln(β/T_P²) = (−E_a/RT_P) + ln(k₀) (1)

where k₀ is a constant. The plot of ln(β/T_P²) vs. 1/T_P has a negative slope. Here, we have used 2, 5, 10 and 15 °C min⁻¹ DSC heating rates for calculating the desorption activation energy of the samples. The corresponding ln(β/T_P²) vs. 1/T_P plot of ball-milled MgH₂ and MgH₂ + 5 wt% G-Fe is shown in Fig. 2(c).

The desorption activation energy of ball-milled MgH₂ (w/o additive) and 5 wt% G-Fe admixed MgH₂ determined from the DSC curves using the Kissinger method is 157.4 and 119.1 kJ mol⁻¹. For pristine MgH₂, it is ~187 kJ mol⁻¹.^16,17Table 1 gives the summary of the calculated apparent desorption activation energy of MgH₂ determined from DSC curves using the Kissinger equation.

Table 1 Summary of desorption activation energy determined for MgH₂ using the Kissinger method

	Pristine MgH2	5 hours ball-milled MgH2	MgH2 + G-Fe (5 wt%)
Activation energy (kJ mol^−1)	~187 (ref. 16 and 17)	157.4	119.1

---

It should be noted that 1 hour ball-milled MgH₂ + 5 wt% graphene nanosheets (without metal decoration) shows the peak desorption temperature at 358 °C for the DSC heating rate of 5 °C min⁻¹.¹⁸ Meanwhile, in the present case, 1 hour ball-milled MgH₂ + 5 wt% G-Fe shows the peak desorption temperature at 323 °C for the DSC heating rate of 5 °C min⁻¹ (refer to Fig. 2(b)). Thus the presence of Fe nanoclusters on graphene greatly lowers the dissociation temperature of MgH₂.

Fig. 3 presents the isothermal reabsorption kinetic curves of dehydrogenated MgH₂ + 5 wt% G-Fe determined at different temperatures (200, 280 and 300 °C) and 20 atm hydrogen pressure. It has been observed that the dehydrogenated MgH₂ + G-Fe sample shows fast absorption kinetics. The sample absorbed more than 5 wt% hydrogen in less than 4 minutes at 300 °C and 20 atm. Meanwhile, the ball-milled (un-catalyzed) dehydrogenated MgH₂ sample took nearly 20 min to absorb 5 wt% hydrogen at 300 °C and 20 atm. This reveals that G-Fe exhibits superior catalytic effect on MgH₂ for both hydrogen release and uptake.

Fig. 3 Reabsorption kinetic curves of dehydrogenated MgH₂ + G-Fe.

Fig. 4(a) & (b) represent the PCI isotherms and Van't Hoff plot of MgH₂ + G-Fe, respectively. From the Van't Hoff plot (Fig. 4(b)), the formation enthalpy and entropy values of MgH₂ with 5 wt% G-Fe were derived. The change in enthalpy and entropy values is −50.4 ± 2.9 kJ per mole of H₂ and 99.8 ± 5.2 J K⁻¹ per mole of H₂, respectively, for the formation of MgH₂. It should be mentioned that the change in formation enthalpy and entropy values for as-received bulk MgH₂ is −74.06 ± 0.42 kJ per mole of H₂ and 133.4 ± 0.7 J K⁻¹ per mole of H₂, respectively.¹⁹ However, the change in formation enthalpy and entropy of MgH₂ was reported for element-coated Mg. For Fe-coated Mg, the formation enthalpy and entropy of MgH₂ were −59.9 ± 1.9 kJ per mole of H₂ and 112.3 ± 3.1 J K⁻¹ per mole of H₂, respectively.²⁰

Fig. 4 (a) PCI desorption and absorption curves of MgH₂ + G-Fe determined at different temperatures and (b) Van't Hoff plot of MgH₂ + G-Fe.

Zhao-Karger et al.²¹ reported that MgH₂ nanoparticles (<3 nm) confined in an activated carbon fiber scaffold system exhibit reduced enthalpy and entropy values of −63.8 kJ per mole of H₂ and 117.2 J K⁻¹ per mole of H₂, respectively. Jia et al.²² observed that MgH₂ nanoparticles with ordered mesoporous carbon, CMK-3, exhibit a reduced enthalpy and entropy of −55.4 kJ per mole of H₂ and 105.7 J K⁻¹ per mole of H₂, respectively. Recently, reduced enthalpy and entropy values of −45 ± 3 kJ per mole of H₂ and 84 ± 5 J K⁻¹ per mole of H₂, respectively, were observed for MgH₂-Ti nanocomposites produced by spark discharge.¹²

It has been pointed out that the thermodynamic properties of the Mg-to-MgH₂ hydrogenation reaction can be dramatically improved at the nanoscale due to the dominant contribution of the high surface to volume ratio.¹² In accordance with previous reports, it is known that Fe nanoparticles²³ and graphene^18,24 are very good catalysts for the MgH₂ system. Therefore, in the present investigation, the observation of low reaction enthalpy and entropy values for the hydrogenation reaction of Mg is due to the nanosize and intermixed character of the disordered graphene and Fe nanoparticles. The presence of Fe nanoparticles and graphene is expected to have a synergetic effect in improving the dehydrogenation behavior and in enhancing the hydrogenation of Mg to MgH₂ through the spillover effect.^25,26

Fig. 5(a) shows the XRD patterns of as-prepared MgH₂ + 5 wt% G-Fe, Fig. 5(b) presents the XRD pattern of dehydrogenated MgH₂ + G-Fe (5 wt%) after five rehydrogenation cycles and Fig. 5(c) shows its XRD pattern after six rehydrogenation cycles.

Fig. 5 Powder X-ray diffraction patterns of MgH₂ + G-Fe: (a) as-prepared, (b) dehydrogenated after five hydrogenation cycles and (c) after six hydrogenation cycles.

It was found that along with the main MgH₂ phase, the MgO phase is detected. It was noticed that after six dehydrogenation and rehydrogenation cycles, the diffraction peak intensity increases, suggesting the grain growth of MgH₂. Assuming that the MgH₂ particles were spherical, the size of the grains were calculated from the XRD peaks using Scherrer's equation.²⁷ This suggests that the average grain size of as-prepared MgH₂ + G-Fe (Fig. 5(a)) is 19 nm and after six cycles (Fig. 5(c)), the grain size of MgH₂ increases to 34.8 nm. Liu et al.²⁴ observed that after six hydrogenation cycles, the grain size of pure-MgH₂ (w/o additives) increased three times (from 10.8 nm before cycling to 32 nm after cycles).

The nucleation and growth of the Mg₂FeH₆ phase was observed in the cycled MgH₂ + G-Fe sample, due to the reaction between elemental Mg and Fe during the hydrogenation process, suggesting that the Fe particles were not free during hydrogenation. However, during dehydrogenation, Mg₂FeH₆ completely transformed into elemental Mg and Fe, benefiting the initial stage of the rehydrogenation process.²⁸

Microstructural analysis of the sample was carried out using TEM. Fig. 6 presents the TEM microstructures of the as-synthesized graphene sheets decorated with Fe nanoclusters (Fig. 6(a)), MgH₂ ball-milled for 5 hours (Fig. 6(b)), MgH₂ + 5 wt% G-Fe (Fig. 6(c)) and MgH₂ + 5 wt% G-Fe after five dehydrogenation/rehydrogenation cycles (Fig. 6(d)).

Fig. 6 TEM microstructure of (a) as-synthesized G-Fe, (b) MgH₂ ball-milled for 5 hours, (c) as-prepared MgH₂ + G-Fe and (d) MgH₂ + G-Fe after six dehydrogenation and rehydrogenation cycles. The insets show the corresponding SAED patterns.

It was found from the TEM micrograph (Fig. 6(a)) that the Fe nanoparticles (indicated by blue arrows) in G-Fe were distributed throughout the graphene surface. The size of Fe nanoparticles ranges from 5 to 15 nm. The hexagonal pattern observed in the SAED pattern (inset of Fig. 6(a)) confirms that the synthesized sample is graphene. EDAX spectral analysis confirms that the composition of Fe nanoparticles in graphene is 2 wt%. Fig. S1 and S2† show the representative Raman and energy dispersive X-ray (EDX) spectra of as-prepared G-Fe. The distribution of graphene sheets (shown by white arrows) can be readily distinguished in the as-prepared and cycled MgH₂ + G-Fe samples (Fig. 6(c) & (d)). The corresponding SAED patterns (insets in Fig. 6(c) & (d)) show the fine grain distribution of the samples. The ring patterns arise in the SAED patterns due to the presence of Fe nanoparticles and the bright patterns are from the MgH₂ particles. It has been observed from the TEM micrograph of the cycled sample that graphene is intercalated between MgH₂ particles, possibly reducing the agglomeration of MgH₂ particles during cycling. In accordance with earlier studies by other researchers, it has been observed that the presence of graphene reduces the grain growth of MgH₂, suggesting that graphene is a very good crystal growth inhibitor during dehydrogenation/rehydrogenation cycles of MgH₂.^25,26

In order to further understand the synergetic effect of Fe nanoclusters and graphene, we have done ab initio studies which are described below.

Computational method

The Grid-based Projector-Augmented Wave method (GPAW) is implemented in density functional theory calculations.²⁹ The grid spacing is set to 0.20 Å with 0.00 eV of smearing. The Γ point is implemented for the sampling of the Brillouin zone. Exchange-correlation of Perdew–Burke–Ernzerhof (PBE) with spin polarization calculations is applied for all calculations.³⁰ Charge transfer analysis is performed using Bader charge analysis.^31,32

The adsorption energies of the MgH₂ cluster and one H atom are calculated using eqn (2)

E_ad = E_{[Surface+adsorbate]} − E_[Surface] − E_[Adsorbate] (2)

The reaction between Fe-decorated graphene and MgH₂ takes place by using the mechanical ball milling process in the experiment. In general, the ball milling process destroys the structure of the material. Hence, atomic clusters are implemented in order to consider the complicated unusual structures. The MgH₂ cluster is used in order to understand the fundamental dehydrogenation properties over the Fe-decorated catalysts as MgH₂ clusters have been shown to be an effective model to understand the hydrogenation and dehydrogenation properties of MgH₂ systems.¹⁰ In addition, graphene within non-periodic boundary conditions is considered. In particular, asymmetric graphene consisting of 32 carbon atoms, shown in Fig. 7(a), is constructed within non-periodic boundary conditions.

Fig. 7 (a) Asymmetric graphene, (b) MgH₂ cluster on symmetric graphene, (c) MgH₂ cluster on Fe-doped asymmetric graphene, (d) MgH₂ on single defected graphene, (e) MgH₂ cluster on large defected graphene, and (f) MgH₂ cluster on large defected graphene with Fe. Interatomic distances of H–C, Mg–C, Mg–H, Mg–Fe, and Fe–H are shown in Å. Atomic color codes: Mg, green; H, white; C, gray; Fe, orange.

Computational results

The MgH₂ cluster is adsorbed on non-defected graphene with/without an Fe atom, single defected graphene, Fe-doped graphene and large graphene with/without an Fe atom. The adsorption energies of the MgH₂ cluster and H atom over each graphene model are collected in Table 2. The structural models are shown in Fig. 7(a–f). The different sizes of defects and the adsorption of Fe monomers are considered based on the graphene in Fig. 7(a).

Table 2 MgH₂ adsorption energy E(MgH₂) and hydrogen desorption energy E(H) in eV over various graphene conditions

	E MgH2	E H
Non-defected graphene	−4.21	−0.55
Non-defected graphene + Fe	−1.36	−0.06
Single defected graphene	−5.38	−0.83
Large defected graphene	−7.89	−2.08
Large defected graphene + Fe	−2.48	−0.53

---

The MgH₂ cluster is adsorbed on the edge of non-defected graphene where one of the H atoms is dissociated and adsorbed on the C atom. The adsorption energy of the MgH₂ cluster and the H atom on non-defected graphene is calculated to be −4.21 eV and −0.55 eV, respectively. Upon the introduction of Fe atom onto non-defected graphene, both H atoms are fully dissociated from MgH₂ clusters, as shown in Fig. 7(c). In particular, it is interesting that the adsorption energy of the MgH₂ cluster and H atom dramatically decreased and was −1.36 eV and 0.06 eV compared to the one without Fe. Thus, a dramatic decrease of H adsorption energy is predicted to allow low temperature dehydrogenation. Single C defect is created at the edge of graphene shown in Fig. 7(d). Both H atoms are dissociated over the single C defected graphene. It is interesting that the adsorption energy of the MgH₂ cluster on single defected graphene is calculated to be −5.38 eV as single defect enhances the MgH₂ cluster adsorption energy by −1.17 eV as compared to the MgH₂ cluster on non-defected graphene. Furthermore, H adsorption energy is calculated to be −0.83 eV where H atoms are adsorbed at the edge of graphene shown in Fig. 7(d). However, a decrease of H adsorption energy is not observed in the single defect case compared to the case of Fe-decorated graphene.

A large defect is created in graphene in order to completely break the structure of graphene where six carbon atoms are taken from graphene consisting of 32 C atoms. The structure of graphene is greatly deformed. The adsorption of MgH₂ cluster on the large defected graphene particularly shows a high adsorption energy of −7.89 eV where both H atoms are dissociated from the MgH₂ cluster, as shown in Fig. 7(e). H adsorption energy is calculated to be −2.08 eV. This also confirms that the edge of the defected graphene is reactive against the MgH₂ cluster. However, high H adsorption energy could require higher temperatures for further dehydrogenation processes. An Fe atom is then adsorbed on the large defected graphene shown in Fig. 7(f). The adsorption energy of the MgH₂ cluster and H atom over the large defected graphene with Fe is dramatically decreased to −2.84 eV and −0.53 eV, respectively, where full dehydrogenation of the MgH₂ cluster is also observed. Thus, one can suggest that the introduction of Fe atom decreases the dehydrogenation temperature of the MgH₂ cluster over defected graphene.

Bader charge analysis is performed for the large defected graphene with/without Fe in order to reveal the physical origin of the effect of Fe. Bader analysis indicates that the Fe atom is positively charged by 0.8 electrons. This indicates that electrons from Fe atoms are transferred to C atoms near Fe atoms. One could suggest that Fe atoms fill and stabilize the defects of graphene. As a result, graphene is less active. On the other hand, electrons in the d-orbital of Fe atoms could enhance the dehydrogenation of the MgH₂ cluster. Thus, the catalytic activities of defected graphene could be controlled by Fe.

Conclusions

Graphene decorated with Fe clusters is experimentally and theoretically proven to be a possible low cost and alternative catalyst towards the hydrogenation reaction compared to transition metal catalysts. Experiments show that graphene decorated with Fe clusters decreases the dehydrogenation temperature of MgH₂ from 322.3 °C to 281.7 °C where the activation energies are experimentally measured to be 119.1 kJ mol⁻¹. Furthermore, the rehydrogenation of MgH₂ with graphene decorated with Fe clusters is dramatically improved where it takes only 4 minutes to absorb 5 wt% hydrogen at 300 °C and 20 atm. After six rehydrogenation cycles, the grain size of MgH₂ increased only 15 nm, showing low order grain growth during cycling. This is advantageous for MgH₂ in maintaining cycling stability, making it suitable for hydrogen storage applications. TEM analysis confirms the low order grain growth of MgH₂ and shows that the graphene decorated with Fe clusters behaves like a shell while MgH₂ is its core. Density functional theory calculations reveal that the location of the active site – the defect in graphene – and the presence of iron clusters at the defect site of graphene are the key factors for the dehydrogenation of MgH₂, as the Fe clusters reduce the adsorption of dissociated H atoms, thus resulting in low-temperature hydrogen release. This suggests that other transition metals could also be effective if charge transfers from the transition metal to graphene are involved. Thus, graphene decorated with Fe clusters could be a potential alternative for transition metal catalysts for hydrogenation reactions and the possibility of utilizing graphene decorated with other metal clusters could further be explored.

Acknowledgements

The authors acknowledge the Hydrogen Energy Centre of Banaras Hindu University for providing the necessary experimental facilities. Financial support from the Department of Science and Technology (IFA12-PH-18) and Ministry of New and Renewable Energy (Mission Mode Project) is acknowledged. Thanks are also due to CMPDI, Ministry of Coal, India. One of the authors, Seema Awasthi, acknowledges the University Grants Commission for financial support under the scheme, Dr. DS Kothari PDF grant (no. F.4-2/2006(BSR)/13-768/2012(BSR)). CPU time is funded by the Japan Society for the Promotion of Science. The calculations are performed at Hokkaido University's academic cloud, Information Initiative Center.

Notes and references

1. V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, N. Kim, K. C. Kemp, P. Hobza, R. Zboril and K. S. Kim, Functionalization of graphene: covalent and non-covalent approaches, derivatives and applications, Chem. Rev., 2012, 112(11), 6156–6214.
2. J. Pyun, Graphene Oxide as Catalyst: Application of Carbon Materials beyond Nanotechnology, Angew. Chem., Int. Ed., 2011, 50(1), 46–48.
3. C. Su, M. Acik, K. Takai, J. Lu, S.-J. Hao, Y. Zheng, P. Wu, Q. Bao, T. Enoki, Y. J. Chabal and K. P. Loh, Probing the catalytic activity of porous graphene oxide and the origin of this behavior, Nat. Commun., 2012, 3, 1298.
4. Y. H. Lu, M. Zhou, C. Zhang and Y. P. Feng, Metal-Embedded Graphene: A Possible Catalyst with High Activity, J. Phys. Chem. C, 2009, 113, 20156–20160.
5. H. R. Byon, J. Suntivich and Y. Shao-Horn, Graphene-based non-noble-metal catalysts for oxygen reduction reaction in acid, Chem. Mater., 2011, 23(15), 3421–3428.
6. S. Sabater, J. A. Mata and E. Peris, Catalyst Enhancement and Recyclability by Immobilization of Metal Complexes onto Graphene Surface by Noncovalent Interactions, ACS Catal., 2014, 4(6), 2038–2047.
7. K. Takahashi, Y. Wang, S. Chiba, Y. Nakagawa, S. Isobe and S. Ohnuki, Low temperature hydrogenation of iron nanoparticles on graphene, Sci. Rep., 2014, 4, 4598.
8. M. S. L. Hudson, H. Raghubanshi, S. Awasthi, T. Sadhasivam, A. Bhatnager, S. Simizu, S. G. Sankar and O. N. Srivastava, Hydrogen uptake of reduced graphene oxide and graphene sheets decorated with Fe nanoclusters, Int. J. Hydrogen Energy, 2014, 39(16), 8311–8320.
9. K. Takahashi, S. Isobe and S. Ohnuki, H₂ dissociation over NbO: The first step toward hydrogenation of Mg, Langmuir, 2013, 29(38), 12059–12065.
10. K. Takahashi, S. Isobe and S. Ohnuki, The catalytic effect of Nb, NbO and Nb₂O₅ with different surface planes on dehydrogenation in MgH₂: Density functional theory study, J. Alloys Compd., 2013, 580, S25–S28.
11. G. Barkhordarian, T. Klassen and R. Bormann, Fast hydrogen sorption kinetics of nanocrystalline Mg using Nb₂O₅ as catalyst, Scr. Mater., 2003, 49(3), 213–217.
12. A. Anastasopol, T. V. Pfeiffer, J. Middelkoop, U. Lafont, R. J. Canales-Perez, A. Schmidt-Ott, F. M. Mulder and S. W. Eijt, Reduced enthalpy of metal hydride formation for Mg-Ti nanocomposites produced by spark discharge generation, J. Am. Chem. Soc., 2013, 135(21), 7891–7900.
13. T. Sadhasivam, M. S. L. Hudson, S. K. Pandey, A. Bhatnagar, M. K. Singh, K. Gurunathan and O. N. Srivastava, Effects of nano size mischmetal and its oxide on improving the hydrogen sorption behaviour of MgH₂, Int. J. Hydrogen Energy, 2013, 38(18), 7353–7362.
14. S. K. Pandey, A. Bhatnagar, R. R. Shahi, M. S. L. Hudson, M. K. Singh and O. N. Srivastava, Effect of TiO₂ Nanoparticles on the Hydrogen Sorption Characteristics of Magnesium Hydride, J. Nanosci. Nanotechnol., 2013, 13, 5493–5499.
15. H. E. Kissinger, Reaction Kinetics in Differential Thermal Analysis, Anal. Chem., 1957, 29(11), 1702–1706.
16. M. Pourabdoli, S. Raygan, H. Abdizadeh and D. Uner, Determination of kinetic parameters and hydrogen desorption characteristics of MgH₂-10 wt% (9Ni-2Mg-Y) nano-composite, Int. J. Hydrogen Energy, 2013, 38(27), 11910–11919.
17. L. Zhang, L. Chen, X. Xiao, X. Fan, J. Shao, S. Li, H. Ge and Q. Wang, Fluorographene nanosheets enhanced hydrogen absorption and desorption performances of magnesium hydride, Int. J. Hydrogen Energy, 2014, 39(24), 12715–12726.
18. G. Liu, Y. Wang, C. Xu, F. Qiu, C. An, L. Li, L. Jiao and H. Yuan, Excellent Catalytic Effects of Highly Crumpled Graphene Nanosheets on Hydrogenation/Dehydrogenation of Magnesium Hydride, Nanoscale, 2013, 5, 1074–1081.
19. M. Paskevicius, D. A. Sheppard and C. E. Buckley, Thermodynamic Changes in Mechanochemically Synthesized Magnesium Hydride Nanoparticles, J. Am. Chem. Soc., 2010, 132(14), 5077–5083.
20. W. Liu, E. J. Setijadi and K. F. Aguey-Zinsou, Tuning the Thermodynamic Properties of MgH₂ at the Nanoscale via a Catalyst or Destabilizing Element Coating Strategy, J. Phys. Chem. C, 2014, 118(48), 27781–27792.
21. Z. Zhao-Karger, J. Hu, A. Roth, D. Wang, C. Kübel, W. Lohstroh and M. Fichtner, Altered thermodynamic and kinetic properties of MgH₂ infiltrated in microporous scaffold, Chem. Commun., 2010, 46, 8353–8355.
22. Y. Jia, C. H. Sun, N. Cheng, M. A. Wahab, J. Cui, J. Zou and X. Yao, Destabilization of Mg-H bonding through nano-interfacial confinement by unsaturated carbon for hydrogen desorption from MgH₂, Phys. Chem. Chem. Phys., 2013, 15, 5814–5820.
23. M. Ismail, N. Juahir and N. S. Mustafa, Improved Hydrogen Storage Properties of MgH₂ Co-Doped with FeCl₃ and Carbon Nanotubes, J. Phys. Chem. C, 2014, 118(33), 18878–18883.
24. G. Liu, Y. Wang, L. Jiao and H. Yuan, Understanding the Role of Few-Layer Graphene Nanosheets in Enhancing the Hydrogen Sorption Kinetics of Magnesium Hydride, ACS Appl. Mater. Interfaces, 2014, 6(14), 11038–11046.
25. S. Pevzner, I. Pri-Bar, I. Lutzky, E. Ben-Yehuda, E. Ruse and O. Regev, Carbon Allotropes Accelerate Hydrogenation via Spillover Mechanism, J. Phys. Chem. C, 2014, 118(46), 27164–27169.
26. Y. Lin, F. Ding and B. I. Yakobson, Hydrogen storage by spillover on graphene as a phase nucleation process, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 041402(R).
27. A. L. Patterson, The Scherrer Formula for X-Ray Particle Size Determination, Phys. Rev., 1939, 56, 978.
28. M. Danaie, A. C. Asselli, J. Huot and G. A. Botton, Formation of the Ternary Complex Hydride Mg₂FeH₆ from Magnesium Hydride (β-MgH₂) and Iron: An Electron Microscopy and Energy-Loss Spectroscopy Study, J. Phys. Chem. C, 2012, 116(49), 25701–25714.
29. J. J. Mortensen, L. B. Hansen and K. W. Jacobsen, Real-space grid implementation of the projector augmented wave method, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 71, 035109.
30. J. P. Perdew, K. Burke and M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett., 1996, 77, 3865.
31. E. Sanville, S. D. Kenny, R. Smith and G. Henkelman, Improved grid-based algorithm for Bader charge allocation, J. Comput. Chem., 2007, 28, 899–908.
32. G. Henkelman, A. Arnaldsson and H. A. Jónsson, A fast and robust algorithm for Bader decomposition of charge density, Comput. Mater. Sci., 2006, 36(3), 354–360.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5cy01016k

---