In situ characterization of the decomposition behavior of Mg(BH₄)₂ by X-ray Raman scattering spectroscopy

Abstract

We present an in situ study of the thermal decomposition of Mg(BH₄)₂ in a hydrogen atmosphere of up to 4 bar and up to 500 °C using X-ray Raman scattering spectroscopy at the boron K-edge and the magnesium L_2,3-edges. The combination of the fingerprinting analysis of both edges yields detailed quantitative information on the reaction products during decomposition, an issue of crucial importance in determining whether Mg(BH₄)₂ can be used as a next-generation hydrogen storage material. This work reveals the formation of reaction intermediate(s) at 300 °C, accompanied by a significant hydrogen release without the occurrence of stable boron compounds such as amorphous boron or MgB₁₂H₁₂. At temperatures between 300 °C and 400 °C, further hydrogen release proceeds via the formation of higher boranes and crystalline MgH₂. Above 400 °C, decomposition into the constituting elements takes place. Therefore, at moderate temperatures, Mg(BH₄)₂ is shown to be a promising high-density hydrogen storage material with great potential for reversible energy storage applications.

---

1 Introduction

Zero-emission propulsion is a long standing societal goal that would contribute to restricting the consumption of fossil fuels and limiting the extent of anthropogenically caused climate change.^1,2 Besides energy storage in electrochemical cells, storage of hydrogen as an energy carrier is a promising step toward realizing this dream. However, hydrogen storage, especially for mobile applications, remains challenging.³ Conventional strategies, i.e. liquefying or pressurizing hydrogen, present the challenges that the resulting storage devices are large and heavy, and therefore do not meet the targets set e.g. by the Department of Energy.⁴

Promising alternatives for hydrogen storage include liquid organic hydrogen carriers,⁵ metal organic frameworks,⁶ carbon nanostructures,^7–9 and metal borohydrides.¹⁰ Hydrogen storage in the solid state, especially in metal borohydrides, is desirable, since these materials can offer simultaneous high gravimetric and volumetric hydrogen densities.¹¹ For metal borohydrides – such as LiBH₄, NaBH₄, Mg(BH₄)₂, and Ca(BH₄)₂ – to function as hydrogen storage materials, reversible desorption, i.e. cyclability, is necessary. However, the reversibility of the decomposition reactions of borohydrides can be hindered by unwanted reaction intermediates, such as for example the [B₁₂H₁₂]²⁻-phases in LiBH₄ and NaBH₄. These phases are kinetically stable and thus hinder the reversibility.¹¹ Although it was found that decomposition proceeds in multiple steps and is even partially reversible, the details of the decomposition mechanisms and possible formation of stable boron phases in the intermediate steps is still debated. The amorphous nature of some of the final and intermediate compounds makes the investigation of the important details of the de- and re-hydrogenation mechanisms difficult. Furthermore, these details strongly depend on the experimental conditions and thus need to be investigated in situ.

Mg(BH₄)₂ is one of the most promising metal borohydride hydrogen storage materials due to its moderate enthalpy of decomposition and has been studied widely in the recent years.^12–18 Hydrogen desorption was investigated with a particular focus on the possible formation of closoboranes under different thermal conditions and in different gaseous media such as hydrogen, argon, and helium or under dynamic vacuum conditions resulting in a variety of suggested decomposition pathways.^15,19–23 The formation of MgB₁₂H₁₂, even in the early stages of decomposition, is often inferred by solid-state-nuclear magnetic resonance (NMR) or evidenced indirectly via solution NMR.

The presence of [B₁₂H₁₂]²⁻ would not support the use of Mg(BH₄)₂ for applications in cycleable hydrogen storage devices due to the kinetically unfavorable re-hydrogenation. Hence, full decomposition into MgB₂ is necessary for cycling,¹⁵ which calls for reaction conditions with temperatures that are not suited for portable storage media. However, it was recently shown that Mg(BH₄)₂ decomposes under dynamic vacuum conditions up to 400 °C without the formation of MgB₁₂H₁₂.²³ Here, [B₁₂H₁₂]²⁻ was observed as a part of the polymeric intermediate MgB_xH_y between 265 °C and 300 °C, which first converts into amorphous B, Mg, and MgH₂ at 400 °C and then converts completely into MgB₂ at 500 °C.²³ Other reports proposed a gradual B–H condensation process from [BH₄]⁻ to higher boranes in the decomposition of Mg(BH₄)₂.^15,24 Furthermore, there are also studies suggesting the formation of MgB₁₂H₁₂ species during the dehydrogenation.^19,20 However, the reaction conditions and, especially, the gaseous atmospheres under which those various investigations took place were not the same and, importantly, may not be applicable to practical applications. To improve reversibility and possibly modify the reaction pathways, these must be understood. Different borohydrides decompose along different pathways without a general rule. For some compounds there seems to be an emerging agreement, but the hydrogen release reactions of Mg(BH₄)₂ are still under discussion. So far, there are only a few inconclusive in situ X-ray diffraction (XRD)^18,25,26 and infrared spectroscopy studies²⁶ reported in the literature and a dependence on the presence of impurities remains an open question.^18,25 Hence, experimental in situ studies using complementary methods are highly desireable. We apply X-ray Raman scattering (XRS) spectroscopy at both, the boron K- and magnesium L_2,3-core levels, to analyze the in situ decomposition of Mg(BH₄)₂ under hydrogen atmosphere (up to 4 bar) at temperatures up to 500 °C. XRS provides the unique and powerful capability to investigate these amorphous and/or nanocrystalline products in situ and under well-defined thermodynamic conditions. We find that after the transition from the low temperature α/γ-Mg(BH₄)₂ phase to the high temperature β-phase, significant amounts of intermediate compound(s) form, which, at higher temperatures, decompose into mostly B, MgH₂, and Mg. We quantify the amount of different phases present during the decomposition and discuss possible strategies to identify the intermediate phase(s) that occur(s) at lower temperatures.

2 Experimental

XRS combines the benefits of soft-X-ray spectroscopy and the properties of hard X-rays making it a valuable method for the in situ study of lightweight/low-Z hydrogen storage materials.²⁷ For example, this non-resonant inelastic X-ray scattering technique has already been used to study the B K-edge in MgB₂, BN, C₂B₁₀H₁₂, B₂O₃, and Li₂B₄O₇.^28–32 Likewise, the Mg L_2,3-edges of, for example, elemental Mg have also been studied.^33,34 The use of hard X-rays as a probe allows the in situ study of low-atomic-weight bulk samples under various reaction atmospheres and even at extreme conditions.^35–37 Recently, Miedema et al. have demonstrated the capabilities of in situ XRS spectroscopy to investigate hydrogen desorption in the possible hydrogen storage materials LiBH₄ and LiBH₄/NaBH₄ nanocomposites.^38,39

In this work, XRS spectra were collected at beamline ID20 of the European Synchrotron Radiation Facility. The beam size on the sample was 150 μm × 250 μm (V × H). The spectrometer employed 36 Si(660) analyzer crystals with a mean scattering angle of 50° (used in the case of the B K-edge), and 24 analyzers with a mean scattering angle of 120° (for the Mg L_2,3-edges). This results in momentum transfers of 4.2 ± 0.5 Å⁻¹ (probing mostly p-DOS) and 8.5 ± 0.5 Å⁻¹ (probing a mixture of s-, p-, and d-DOS), respectively. The choice of these different momentum transfers for the two different edges is dictated by the energy-loss position of the dominating Compton profile, which varies as a function of momentum transfer. The core onset extraction from the data is feasible if the peak of the Compton profile is far enough from the energy region of interest; hence in XRS different edges must often be measured with different momentum transfers. The overall energy resolution was 0.6 eV.

We purchased Mg(BH₄)₂ and all reference materials, i.e. amorphous B, H₃BO₃, MgB₂, MgH₂, and MgO, from Sigma-Aldrich and used them without further processing except for the nanoscale MgB₁₂H₁₂/carbon sample, which we prepared by a gas–solid reaction of carbon-supported Mg(BH₄)₂ with a B₂H₆/H₂ atmosphere, as previously discussed.⁴⁰

We used a custom high-temperature/high-pressure sample cell for the in situ measurements, as described in detail elsewhere.⁴¹ In short, we pressed the native Mg(BH₄)₂ powder into a circular, resistively heated sample holder (see inset in Fig. 1(b)) and covered the sample using Be foil (0.1 mm thickness) to prevent sample spillage during annealing. The reaction chamber was closed by a half-spherical Be dome (wall thickness 1.7 mm), which allowed a field of view of 2π for the incident and scattered X-rays. With this cell, we performed measurements at ambient conditions (room temperature/pressure) as well as at temperatures between 200 °C and 500 °C in steps of 100 °C. At each temperature, we measured several XRS spectra first of the of B K-edge and then the Mg L_2,3-edge, checked them for consistency, and averaged signals from several analyzer crystals and different scans. Only the spectra of the boron K-edge measured at 300 °C exhibited a slight time dependence, which we will discuss later. The desired core edges were extracted from the overall spectrum by removing the contribution from valence electrons using parameterized PearsonVII functions. Details of the data analysis can be found elsewhere.^42,43 We normalized all XRS curves to the area below the spectra in the energy loss range from 185–212 eV (45–65 eV) for the B (Mg) edge. In this way, the dependence on the stoichiometry of the sample is not considered; i.e., MgB₁₂H₁₂ has 12 B atoms per formula unit whereas H₃BO₃ only has one B atom per formula unit, and the fit parameters describe the contribution of a single B atom within each compound to the scattering signal. However, since the formula units are known, the stoichiometries can simply be derived, as shown in parentheses in Table 1 alongside the fit parameters. For Mg, the stoichiometric- and fit-values are equivalent since each compound only has one Mg atom per formula unit.

Fig. 1 (a) Hydrogen evolution measured at a constant heating rate of 10 °C min⁻¹ using mass spectrometry (MS) for comparison to in situ XRS data. The colored dots indicate the temperatures chosen for XRS spectra. (b) In situ measurement of gas pressure inside the sample chamber as a function of temperature measured during XRS experiments. The gray shaded areas indicate the regions corresponding to each XRS spectrum. The insets pictures are (top left) the exposed Cu sample holder filled with Mg(BH₄)₂ and covered with a Be window, (along the p–T-curve) tomography images collected just before the in situ XRS measurements at the respective temperatures, and (bottom right) a schematic drawing of a vertical cut through the sample holder and its orientation with respect to incoming X-rays and the detector.

Table 1 Fit parameters and molar fractions (in parentheses) from multi-component analysis of the B K-edge. Dashes (—) indicate that this compound was neglected in the fit. The asterisk indicates that these values are in favor of an occurrence of an intermediate instead of MgB₁₂H₁₂ and B as discussed in the text

T [°C]	MgB12H12	B	β-Mg(BH4)2
300	0.12 ± 0.02 (0.02 ± 0.05)*	0.23 ± 0.02 (0.41 ± 0.06)*	0.64 ± 0.02 (0.57 ± 0.05)
400	0.95 ± 0.02 (0.61 ± 0.08)	0.05 ± 0.02 (0.39 ± 0.14)	—
500	0.71 ± 0.08 (0.17 ± 0.04)	0.29 ± 0.04 (0.83 ± 0.10)	—

---

3 Results and discussion

Fig. 1(a) shows the temperature dependent H₂ evolution of Mg(BH₄)₂ measured by mass spectrometry at a constant heating rate of 10 °C min⁻¹, which was used to find optimized annealing temperatures for the XRS measurement. We chose to measure in situ XRS spectra at the temperatures indicated by colored dots in the image: at 200 °C, before decomposition, at 300 °C to investigate possible intermediate reaction compounds at the initial stage of hydrogen release, and at 400 °C and 500 °C to investigate the solid products remaining after hydrogen release.

The approximately corresponding pressure–temperature relation measured during the in situ XRS experiment is shown in Fig. 1(b). In this case, the sample was heated to the indicated temperature and we waited until the vapor pressure reached an approximate equilibrium. The p–T-regimes during which we recorded XRS spectra are indicated by gray shaded areas in Fig. 1. The pressure increase during heating is a combined effect from both, gas law expansion and H₂ release from Mg(BH₄)₂. During the measurements at above-ambient temperature, we observed continued hydrogen desorption and thus expect further decomposition on much longer time scales. Hence, the measured spectra depict an averaged snapshot of the sample composition in a non-equilibrium state on a timescale of 2 hours rather than the final reaction products at a certain temperature. At each temperature, directly before the XRS measurement, we collected tomographic images of the sample (the details of this technique are reported elsewhere⁴⁴), shown as insets in Fig. 1(b). In the last steps of decomposition, we observed macroscopic cracks in the sample (just before 400 °C and at 500 °C). At these temperatures, the sample pellet's integrity is likely to be compromised due to large volume changes during the formation of Mg and MgO, as discussed later, as well as possible heat and gas evolution.

Fig. 2 shows the in situ spectra of the B K-edge and the Mg L_2,3-edges. The minor differences between the spectra at room temperature and at 200 °C are due to the structural phase transition of α/γ-Mg(BH₄)₂ to β-Mg(BH₄)₂, which occurs between 180 °C and 190 °C.^13,45,46 There is an associated small release of H₂ already at 200 °C, which may be due to the presence of impurities.¹⁸ At temperatures above 200 °C, β-Mg(BH₄)₂ starts to decompose, which is manifested by more dramatic changes in both the B K-edge and the Mg L_2,3-edges (prominent changes are marked by arrows in Fig. 2(a) and (b)).

Fig. 2 XRS spectra of Mg(BH₄)₂ at the (a) B K-edge and (b) Mg L_2,3-edges, measured during thermal decomposition. (c) XRS spectra at the B K-edge of various B-containing reference compounds at room temperature. (d) XRS spectra at the Mg L_2,3-edges of various Mg-containing reference compounds at room temperature.

The in situ XRS spectra can be compared to XRS spectra of reference compounds measured at room temperature to analyze the dehydrogenation reaction pathway of Mg(BH₄)₂. Here, we use a multi-component analysis method to quantify the phases present at each stage during thermal decomposition. We considered various prominent candidates for decomposition and intermediate products: B, a MgB₁₂H₁₂/carbon compound, MgB₂, and metallic Mg, as well as H₃BO₃ and MgO to detect possible sample oxidation. These reference spectra are shown in Fig. 2(c) and (d). The XRS spectrum of metallic Mg was previously reported³⁴ and all other spectra were measured in this work. The presence of numerous unique features in each reference spectrum allow each to be used as fingerprints to identify the main decomposition products of native Mg(BH₄)₂ by a multi-component fitting method. The MgB₁₂H₁₂/carbon-reference is representative of a wide variety of higher Mg boranes (Mg_xB_yH_z) with a similar local chemical environment of B and Mg.

The multi-component fit of the B K-edge was performed by considering contributions from B and MgB₁₂H₁₂, except for the initial decomposition at 300 °C where, in addition, we also considered the β-Mg(BH₄)₂ phase (spectrum measured at 200 °C). Occurence of β-Mg(BH₄)₂ for higher temperatures and occurence of MgB₂ in the whole temperature range could be excluded due to the analysis of the Mg L-edge as discussed later. The possibility of the oxidation of boron during annealing was excluded due to the lack of any feature at 194 eV (compare e.g. Miedema et al.³⁸) and hence H₃BO₃ was not a component in the fit. The Mg L_2,3-edges were fitted with contributions from all six Mg-containing references: MgB₁₂H₁₂, metallic Mg, MgH₂, MgB₂, MgO, and the β-Mg(BH₄)₂ phase. The best fit results are shown in Fig. 3 and the resulting best fit parameters are given in Tables 1 and 2.

Fig. 3 XRS spectra of Mg(BH₄)₂ at the (a) B K-edge and (b) Mg L_2,3-edges at different temperatures during thermal decomposition; experimental results (colored lines) and multi-component fits (black dashed and solid lines). The spectra are plotted with vertically offset for clarity.

Table 2 Fit parameters (i.e molar fractions) from multi-component analysis of the Mg L_2,3-edge. The asterisk indicates values that favor the occurrence of an intermediate rather than MgB₁₂H₁₂ as discussed in the text

T [°C]	MgB12H12	MgH2	Metallic Mg	MgO	β-Mg(BH4)2
300	(0.32 ± 0.05)*	0.28 ± 0.05	0.01 + 0.04	0.00 + 0.04	0.35 ± 0.02
400	0.29 ± 0.05	0.50 ± 0.01	0.15 ± 0.03	0.16 ± 0.04	0.00 + 0.04
500	0.23 ± 0.05	0.35 ± 0.02	0.21 ± 0.01	0.27 ± 0.02	0.00 + 0.04

---

B K-edge

A representative fit of the B K-edge at 300 °C could be achieved only when β-Mg(BH₄)₂ is considered as a reference for this temperature as well. Analyzing the time dependence of the B K-edge spectra taken at 300 °C we find a decreasing amount of this phase with annealing on timescales below 1 h. The amount of MgB₁₂H₁₂ and B estimated from this fit at 300 °C is far from resembling the stoichiometry of the starting material, indicating that none of theses compounds is a major constituent of the sample a this temperature. We conclude that after transition to the high-temperature phase of Mg(BH₄)₂ at 200 °C, the sample starts to decompose when heated to 300 °C and hydrogen gas is released. This implies the absence of both MgB₁₂H₁₂ and solid B in the decomposition product, and points to the formation of an unknown intermediate phase. This intermediate phase has been identified as Mg(B₃H₈)₂ in other work.^23,24

We find a good agreement between the measured spectrum at 400 °C and a fit involving our MgB₁₂H₁₂ reference sample and small amounts of B. At 500 °C, the spectral weight of B increases significantly at the expense of the MgB₁₂H₁₂ contribution, suggesting that some of the MgB₁₂H₁₂-like phase decomposes into elemental B and Mg during the observed H₂ release. MgB₁₂H₁₂ has been found to decompose at temperatures above 400 °C into B and Mg.⁴⁰ At all temperatures explored in this work, the occurrence of MgB₂ could be ruled out due to its prominent pre-edge feature at 187 eV. This is in contrast to earlier studies employing vacuum conditions.^14,23 We note that the present results characterize the non-equilibrium composition of the sample upon annealing for times on the order of 2 h and might not describe the final composition after long-term annealing at the same temperatures. Consequently, the results obtained here should be used to identify decomposition reaction pathways including amorphous reaction products, but not necessarily to identify equilibrium states.

The MgB₁₂H₁₂ carbon-nanocomposite reference material, has been used in this work to identify contributions from MgB₁₂H₁₂-like species, including MgB₁₂H₁₂ (as suggested in other work¹⁵), B₁₂H₁₂ dimers, or larger borane polymers interconnected by Mg atoms as they may show similar Mg and B fingerprints due to their similar chemical environment.⁴⁰ However, assuming MgB₁₂H₁₂ to be a decomposition product according to the fit of the B K-edge at 400 °C and 500 °C, results in a larger Mg : B ratio than expected from the 1 : 2 stoichiometry of the starting material which will be discussed after evaluation of the Mg L-edges.

Mg L_2,3-edges

The high-temperature phase of Mg(BH₄)₂ decomposes upon annealing at 300 °C. The corresponding fit gives similar contributions of MgH₂ and β-Mg(BH₄)₂. Although we find signatures of our MgB₁₂H₁₂ reference spectrum in the in situ data at 300 °C, the occurrence of MgB₁₂H₁₂ can be ruled out at this temperature based on the B K-edge analysis. Also, the mismatch of MgB₁₂H₁₂ : Mg(BH₄)₂ ratios extracted from both edges independently may be due to the proceeding decomposition of Mg(BH₄)₂ during the measurements at 300 °C. This finding is consistent with the presence of an unknown intermediate phase and the formation of MgH₂.

This intermediate phase then decomposes into a MgB₁₂H₁₂-like phase (as discussed above), MgH₂, metallic Mg, and small amounts of B (as inferred from the B K-edge fit) between 300 °C and 400 °C. In addition, a contribution from MgO was observed at high temperatures as indicated by the spectral feature at 57 eV energy loss. Most probably, part of the metallic Mg decomposition product was oxidized by residual oxygen in the reaction chamber. Elemental Mg is clearly present in the sample at this temperature because of the small pre-edge shoulder at around 51 eV in the Mg L_2,3-edges. Between 400 °C and 500 °C, the spectral weights of MgB₁₂H₁₂ and MgH₂ decrease, whereas those of metallic Mg and – correspondingly – of MgO increase. We infer from this that MgB₁₂H₁₂-like species as well as MgH₂ decompose into elemental Mg and B accompanied by hydrogen release (also reported elsewhere⁴⁰). The increasing contribution of Mg is coupled to an increase in MgO content. The decrease of MgB₁₂H₁₂ and MgH₂ content directly converts to an increase of Mg and MgO content within the errors of the fit. Deviations of the fits from the measured spectra may be due to the fact that we used a nanocomposite MgB₁₂H₁₂ sample as a reference. Comparing the fit results obtained for the Mg L_2,3- and B K-edges, similar amounts of decomposed MgB₁₂H₁₂ (or MgB₁₂H₁₂-like phases) are found. It should be noted, that at 500 °C contributions from MgB₁₂H₁₂-like species and MgH₂ are still present in the sample, suggesting that the decomposition reaction is slow and not fully completed. Furthermore, the stoichiometric analysis of all Mg- and B-containing phases based on the fits of both edges yields an Mg : B ratio of approximately 1 : 4 if MgB₁₂H₁₂ is assumed to be the decomposition product. This ratio deviates significantly from the 1 : 2 ratio of the native Mg(BH₄)₂, which implies that instead of MgB₁₂H₁₂, a polymer or other intermediate is present with a stoichiometry of 1 : 6 (Mg : B) that exhibits a similar local Mg and B environment compared to MgB₁₂H₁₂.

Decomposition pathway

The information obtained in this work from both the B K-edge and the Mg L_2,3-edges gives a detailed picture of the thermal decomposition of Mg(BH₄)₂. It must be emphasized that in this current case, the results obtained give in situ snapshots of the sample during the decomposition process rather than information about equilibrium states. The high temperature phase of Mg(BH₄)₂ decomposes into MgH₂ and an intermediate phase or a mixture of intermediates which cannot be further identified at 300 °C. This phase then transforms at 400 °C into MgH₂, Mg, and elemental B, either via the formation of MgB₁₂H₁₂-like species with smaller Mg to B ratio and subsequent decomposition or via a polymerization process as described in other work.^23,24 At 500 °C MgH₂ decomposes into elemental Mg. All phases of the decomposition are accompanied by the release of hydrogen gas (Fig. 4).

Fig. 4 Schematic summary of the thermal decomposition of Mg(BH₄)₂.

The present study shows that the intermediate formed in the main decomposition step of Mg(BH₄)₂ (between 200 °C and 300 °C) is not MgB₁₂H₁₂ and supports earlier studies,²³ namely that an intermediate with Mg_xB_yH_z stoichiometry forms first, together with MgH₂. This intermediate might be Mg(B₃H₈)₂ as identified in these earlier studies^23,24 which would be consistent with the estimated stoichiometry using our fit results. At higher temperatures, MgH₂ decomposes into Mg and H₂, while the intermediate Mg_xB_yH_z may react under the release of hydrogen to growing agglomerates containing B₁₂H₁₂-like units until, finally, all hydrogen is released. Between 300 °C and 400 °C a significant amount of hydrogen release can be achieved without the occurrence of unwanted byproducts such as MgB₁₂H₁₂ and elemental B. It is unknown from this work whether the reaction, if stopped at 300 °C, would proceed further given more time.

In order to identify the intermediate products in the early stages of the decomposition by XRS, it would be desirable to measure the reference spectra of possible intermediate products discussed elsewhere in the literature (e.g.ref. 47 and 48) and incorporate them into the multi-component fitting analysis. However, many of these possible intermediate structures were predicted via theoretical modeling and are difficult to synthesize or are unstable in freestanding form. Nevertheless, experimental XRS spectra may provide a unique benchmark for calculated XRS spectra both at the Mg L_2,3- and B K-edges of the proposed intermediates.

4 Summary and outlook

We have presented an in situ XRS spectroscopy study of the thermal decomposition of Mg(BH₄)₂ to understand the reaction pathway during hydrogen desorption. From the XRS spectra we could infer quantitative information even on amorphous products during the decomposition of borohydrides. Our work opens new perspectives for future investigations, e.g. of nanoconfined samples, layer-protected hydrides, and eutectic mixtures. Spectral fingerprinting via multi-component fits of measured reference compounds to the in situ spectra of the decomposing Mg(BH₄)₂ were used to identify the major decomposition products. Consistent information from the B K-edge and the Mg L_2,3-edges suggests the formation of yet to be identified intermediate decomposition products at temperatures up to 300 °C and the subsequent decomposition of these intermediates into a phase with an Mg : B stoichometry of 1 : 6 and MgB₁₂H₁₂-like spectral shape as well as MgH₂ and elemental B at 400 °C. Further annealing yields mainly B and Mg. XRS has proven to be a valuable tool to infer quantitative information on even amorphous products during the decomposition reaction and opens new possibilities for future investigations. Future theoretical approaches may yield information about the intermediate compounds. During the decomposition of Mg(BH₄)₂ at 300 °C, we observe a significant release of hydrogen gas without the formation of stable B phases, which highlights the great promise of Mg(BH₄)₂ as a reversible hydrogen storage medium.

Acknowledgements

We acknowledge Harald Müller and Christian Henriquet for expert help and assistance during the experiment. We thank the European Synchrotron Radiation Facility for providing synchrotron radiation and the technical and scientific staff of the ESRF beamline ID20 for support and valuable discussions. This work was supported by the Academy of Finland (Grants 1254065, 1283136, and 1259526) and the University of Helsinki Research Funds (Grants 490076 and 490064). Y. Yan and A. Remhof would like to acknowledge the financial support granted by Switzerland through the Swiss contribution to the enlarged European Union.

References

1. A. J. Churchard, E. Banach, A. Borgschulte, R. Caputo, J.-C. Chen, D. Clary, K. J. Fijalkowski, H. Geerlings, R. V. Genova and W. Grochala, et al. , Phys. Chem. Chem. Phys., 2011, 13, 16955–16972.
2. C. Liu, F. Li, L.-P. Ma and H.-M. Cheng, Adv. Mater., 2010, 22, E28–E62.
3. L. Schlapbach and A. Züttel, Nature, 2001, 414, 353–358.
4. DOE Targets for Onboard Hydrogen Storage Systems for Light-Duty Vehicles.
5. D. Teichmann, W. Arlt, P. Wasserscheid and R. Freymann, Energy Environ. Sci., 2011, 4, 2767–2773.
6. N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim, M. O'Keeffe and O. M. Yaghi, Science, 2003, 300, 1127–1129.
7. B. Panella, M. Hirscher and S. Roth, Carbon, 2005, 43, 2209–2214.
8. Z. Yang, Y. Xia and R. Mokaya, J. Am. Chem. Soc., 2007, 129, 1673–1679.
9. N. P. Stadie, J. J. Vajo, R. W. Cumberland, A. A. Wilson, C. C. Ahn and B. Fultz, Langmuir, 2012, 28, 10057–10063.
10. S.-i. Orimo, Y. Nakamori, J. R. Eliseo, A. Züttel and C. M. Jensen, Chem. Rev., 2007, 107, 4111–4132.
11. H.-W. Li, Y. Yan, S.-i. Orimo, A. Züttel and C. M. Jensen, Energies, 2011, 4, 185–214.
12. R. Černý, Y. Filinchuk, H. Hagemann and K. Yvon, Angew. Chem., 2007, 119, 5867–5869.
13. J.-H. Her, P. W. Stephens, Y. Gao, G. L. Soloveichik, J. Rijssenbeek, M. Andrus and J.-C. Zhao, Acta Crystallogr., Sect. B: Struct. Sci., 2007, 63, 561–568.
14. N. Hanada, K. Chłopek, C. Frommen, W. Lohstroh and M. Fichtner, J. Mater. Chem., 2008, 18, 2611–2614.
15. G. L. Soloveichik, Y. Gao, J. Rijssenbeek, M. Andrus, S. Kniajanski, R. C. Bowman, S.-J. Hwang and J.-C. Zhao, Int. J. Hydrogen Energy, 2009, 34, 916–928.
16. Y. Yan, Y. S. Au, D. Rentsch, A. Remhof, P. E. de Jongh and A. Züttel, J. Mater. Chem. A, 2013, 1, 11177–11183.
17. Y. S. Au, Y. Yan, K. P. De Jong, A. Remhof and P. E. De Jongh, J. Phys. Chem. C, 2014, 118, 20832–20839.
18. N. P. Stadie, E. Callini, B. Richter, T. R. Jensen, A. Borgschulte and A. Züttel, J. Am. Chem. Soc., 2014, 136, 8181–8184.
19. S.-J. Hwang, R. C. Bowman, J. W. Reiter, J. Rijssenbeek, G. L. Soloveichik, J.-C. Zhao, H. Kabbour and C. C. Ahn, J. Phys. Chem. C, 2008, 112, 3164–3169.
20. H. Li, K. Miwa, N. Ohba, T. Fujita, T. Sato, Y. Yan, S. Towata, M. Chen and S. Orimo, Nanotechnology, 2009, 20, 204013.
21. G. Severa, E. Rönnebro and C. M. Jensen, Chem. Commun., 2010, 46, 421–423.
22. S. Gupta, I. Z. Hlova, T. Kobayashi, R. V. Denys, F. Chen, I. Y. Zavaliy, M. Pruski and V. K. Pecharsky, Chem. Commun., 2013, 49, 828–830.
23. Y. Yan, A. Remhof, D. Rentsch and A. Züttel, Chem. Commun., 2015, 51, 700–702.
24. M. Chong, A. Karkamkar, T. Autrey, S.-i. Orimo, S. Jalisatgi and C. M. Jensen, Chem. Commun., 2011, 47, 1330–1332.
25. K. Chłopek, C. Frommen, A. Léon, O. Zabara and M. Fichtner, J. Mater. Chem., 2007, 17, 3496–3503.
26. M. Paskevicius, M. B. Ley, D. A. Sheppard, T. R. Jensen and C. E. Buckley, Phys. Chem. Chem. Phys., 2013, 15, 19774–19789.
27. W. Schülke, Electron dynamics by inelastic X-ray scattering, OUP Oxford, 2007.
28. A. Mattila, J. Soininen, S. Galambosi, S. Huotari, G. Vankó, N. Zhigadlo, J. Karpinski and K. Hämäläinen, Phys. Rev. Lett., 2005, 94, 247003.
29. S. K. Lee, P. J. Eng, H.-k. Mao, Y. Meng, M. Newville, M. Y. Hu and J. Shu, Nat. Mater., 2005, 4, 851–854.
30. S. K. Lee, P. J. Eng, H.-k. Mao, Y. Meng and J. Shu, Phys. Rev. Lett., 2007, 98, 105502.
31. Y. Feng, J. Soininen, A. Ankudinov, J. Cross, G. Seidler, A. Macrander, J. Rehr and E. Shirley, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 165202.
32. T. T. Fister, F. D. Vila, G. T. Seidler, L. Svec, J. C. Linehan and J. O. Cross, J. Am. Chem. Soc., 2008, 130, 925–932.
33. T. Fister, G. Seidler, L. Wharton, A. Battle, T. Ellis, J. Cross, A. Macrander, W. Elam, T. Tyson and Q. Qian, Rev. Sci. Instrum., 2006, 77, 63901.
34. H. Sternemann, J. Soininen, C. Sternemann, K. Hämäläinen and M. Tolan, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 075118.
35. S. K. Lee, J.-F. Lin, Y. Q. Cai, N. Hiraoka, P. J. Eng, T. Okuchi, H.-k. Mao, Y. Meng, M. Y. Hu and P. Chow, et al. , Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 7925–7929.
36. C. J. Sahle, C. Sternemann, C. Schmidt, S. Lehtola, S. Jahn, L. Simonelli, S. Huotari, M. Hakala, T. Pylkkänen and A. Nyrow, et al. , Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 6301–6306.
37. J. Inkinen, A. Sakko, K. Ruotsalainen, T. Pylkkänen, J. Niskanen, S. Galambosi, M. Hakala, G. Monaco, S. Huotari and K. Hämäläinen, Phys. Chem. Chem. Phys., 2013, 15, 9231–9238.
38. P. S. Miedema, P. Ngene, A. M. van der Eerden, T.-C. Weng, D. Nordlund, D. Sokaras, R. Alonso-Mori, A. Juhin, P. E. de Jongh and F. M. de Groot, Phys. Chem. Chem. Phys., 2012, 14, 5581–5587.
39. P. S. Miedema, P. Ngene, A. M. Van Der Eerden, D. Sokaras, T.-C. Weng, D. Nordlund, Y. S. Au and F. M. De Groot, Phys. Chem. Chem. Phys., 2014, 16, 22651–22658.
40. A. Remhof, Y. Yan, D. Rentsch, A. Borgschulte, C. M. Jensen and A. Züttel, J. Mater. Chem. A, 2014, 2, 7244–7249.
41. P. Mauron, M. Bielmann, A. Remhof and A. Züttel, Rev. Sci. Instrum., 2011, 82, 065108.
42. H. Sternemann, C. Sternemann, G. Seidler, T. Fister, A. Sakko and M. Tolan, J. Synchrotron Radiat., 2008, 15, 162–169.
43. C. J. Sahle, A. Mirone, J. Niskanen, J. Inkinen, M. Krisch and S. Huotari, J. Synchrotron Radiat., 2015, 22, 400–409.
44. S. Huotari, T. Pylkkänen, R. Verbeni, G. Monaco and K. Hämäläinen, Nat. Mater., 2011, 10, 489–493.
45. Y. Filinchuk, R. Cerny and H. Hagemann, Chem. Mater., 2009, 21, 925–933.
46. Y. Filinchuk, B. Richter, T. R. Jensen, V. Dmitriev, D. Chernyshov and H. Hagemann, Angew. Chem., Int. Ed., 2011, 50, 11162–11166.
47. V. Ozolins, E. Majzoub and C. Wolverton, J. Am. Chem. Soc., 2008, 131, 230–237.
48. Y. Zhang, E. Majzoub, V. Ozolinš and C. Wolverton, J. Phys. Chem. C, 2012, 116, 10522–10528.

---