Synthesis of LiAlH 4 Nanoparticles Leading to a Single Hydrogen Release Step upon Ti Coating

Lithium aluminum hydride (LiAlH4) is an interesting high capacity hydrogen storage material with fast hydrogen release kinetics when mechanically activated with additives. Herein, we report on a novel approach to produce nanoscale LiAlH4 via a bottom-up synthesis. Upon further coating of these nanoparticles with Ti, the composite nanomaterial was found to decompose at 120 ◦C in one single and extremely sharp exothermic event with instant hydrogen release. This finding implies a significant thermodynamic alteration of the hydrogen properties of LiAlH4 induced by the synergetic effects of the Ti catalytic coating and nanosizing effects. Ultimately, the decomposition path of LiAlH4 was changed to LiAlH4 → Al + LiH + 3/2H2.


Introduction
Hydrogen holds enormous potential to be the ultimate energy carrier of the future [1][2][3][4][5].Yet the lack of an effective method for storing hydrogen with high density currently restrains its widespread utilization.Solid state hydride materials offer promising possibilities to deliver high capacity, safe, and compact hydrogen storage systems [5,6].However, so far no material satisfies all the requirements for practical mobile application [3].For hydrogen storage purposes, LiAlH 4 is one of the most promising and interesting candidates owing to its competitive total hydrogen capacity as well as its low temperature for hydrogen release with fast kinetics [3,6].When doped with catalysts, about 75% of its stored hydrogen can be released below 100 • C in about 1.5 h, while 30 min is required when at 150 • C [7][8][9][10][11].A particularity of LiAlH 4 is the exothermic nature of its initial hydrogen release following the widely accepted Steps 1-3 [12][13][14].
Step 1a 150-175 Step 1a corresponds to the melting of LiAlH 4 quickly followed by the exothermic Step 1b with a reported ∆H of decomposition of −10 kJ•mol −1 H 2 [1,2].Hence, the reversibility of this step is generally believed to be thermodynamically impossible under practical conditions of temperature and pressure [3,4].Indeed, theoretical studies have suggested the need for extreme hydrogen pressures >100 MPa at room temperature for the direct regeneration of LiAlH 4 from Li 3 AlH 6 [2,3,5].In Step 2, Li 3 AlH 6 decomposes via an endothermic reaction with a ∆H of 25 kJ•mol −1 H 2 [6].Hence the regeneration of Li 3 AlH 6 from LiH and Al can be considered thermodynamically feasible.The last decomposition occurring at temperatures higher than 400 • C precludes this step from practical hydrogen storage application, reducing the total hydrogen capacity to 7.9 mass % H 2 (below 300 • C).
Accordingly, current investigations have mainly focused at the destabilization of LiAlH 4 via catalyst doping and/or mechanical milling with additives to further improve its dehydrogenation kinetics and potentially achieve some degree of reversibility.Hence, upon mechanically milling LiAlH 4 with TiCl 3 , ZrCl 4 , or nanosized Ni, lower hydrogen desorption temperatures have been achieved [7][8][9][10][11]; however, with no successful rehydrogenation.Indirect rehydrogenation routes have also been demonstrated through the use of wet synthesis approaches involving the formation of solvent adducts of LiAlH 4 -i.e., THF, Et 2 O, diglyme, and Me 2 O-and in most cases activated/catalyzed Al [12][13][14][15][16].Such an indirect route not only needs to be performed off-board while requiring the additional process of desolvation, but also can compromise the integrity of activated doped LiAlH 4 .For example, for the multi-step regeneration method developed by Graetz et al., and involving the use of THF, successive doping and redoping is required to maintain the same hydrogen storage performance before and after each desolvation of THF [13].Chen et al. achieved a partial direct reversibility of titanium-catalyzed Li 3 AlH 6 , i.e., to reverse Step 2, without the need of solvent desolvation under a hydrogen pressure of 40 MPa only [17].However, the evidence of regenerated Li 3 AlH 6 was a few non-exclusive XRD peaks without structure determination of the rehydrided Li 3 AlH 6 and despite this promising result no follow-up study was reported.
Previous work from our group revealed the possibility of direct rehydrogenation of LiAlH 4 via nanoconfinement in mesoporous carbon and this could be achieved under mild conditions of pressures and temperatures following the assumed route of 3LiH + Al + 3/2H 2 → Li 3 AlH 6 + 2Al + 3H 2 → 3LiAlH 4 [18].Furthermore, positive alternations of hydrogen storage properties were observed with a core-shell method for the individual stabilization of NaBH 4 or LiH nanoparticles avoiding some drawbacks of porous host scaffolds [19,20].This impetus led to the current work which investigated a simple route for the synthesis of isolated LiAlH 4 nanoparticles and their stabilization for hydrogen cycling through a core-shell approach, with Ti acting as a shell facilitating the retention of molten LiAlH 4 upon dehydrogenation as well as a "gateway" for hydrogen absorption/desorption.Ti based additives have been widely studied via top-down ball milling for LiAlH 4 , as well as with other alanate systems e.g., NaAlH 4 and Mg(AlH 4 ) 2 with promising results, i.e., much improved hydrogen storage properties [21][22][23][24][25][26].Indeed, Kojima et al. determined the catalytic effects of various metal chlorides on LiAlH 4 , and revealed their positive effects in terms of hydrogen release in the order of TiCl 3 > ZrCl 4 > VCl 3 > NiCl 2 > ZnCl 2 [27].Amama et al. also investigated the properties of ball milled LiAlH 4 with various dopants, and concluded that Ti-based additives lead to the most enhancement in terms of dehydrogenation kinetics and reduction of the decomposition temperature [9].Based on these previous findings and the potential of a nanosizied approach [28,29], it is thus interesting to investigate the effect of Ti doping/coating on nanosized LiAlH 4 .

Results
In principle, the nanosynthesis of LiAlH 4 can be achieved via common solvent evaporation methods [30].In this procedure, a solution of the nanoparticles' precursor is evaporated under high vacuum [31,32] and this results in the rapid nucleation of the solute materials because of the supersaturated nature of the solution.The growth of the nuclei is then limited via steric repulsion of suitable stabilizers.Currently, the syntheses of most nanoparticles through solvent evaporation are performed in an aqueous solution [33,34].However, for hydride materials which are highly oxidized by water, the synthesis should be performed under non-aqueous environments.To achieve the preparation of nanosized LiAlH 4 particles, we thus used a THF solution of 2M LiAlH 4 with 1-dodecanethiol as a stabilizer and weak capping ligand.
As shown in Figure 1, at the end of the evaporation process the XRD pattern of the material confirmed the preparation of LiAlH 4 with broader diffraction peaks indicating its nanostructure.The nanosized nature of the material was confirmed by TEM analysis showing isolated spherical nanoparticles with a particle size ranging from 2 to 16 nm (Figure 2).preparation of nanosized LiAlH4 particles, we thus used a THF solution of 2M LiAlH4 with 1-dodecanethiol as a stabilizer and weak capping ligand.
As shown in Figure 1, at the end of the evaporation process the XRD pattern of the material confirmed the preparation of LiAlH4 with broader diffraction peaks indicating its nanostructure.The nanosized nature of the material was confirmed by TEM analysis showing isolated spherical nanoparticles with a particle size ranging from 2 to 16 nm (Figure 2).EDS analysis further confirmed the Al content of these nanoparticles that were thus believed to correspond to LiAlH4 (Figure 2f).In comparison, bulk LiAlH4 was imaged as large particles with no defined morphology (Figure 2a), hence the synthesis of nanosized LiAlH4 was successful.preparation of nanosized LiAlH4 particles, we thus used a THF solution of 2M LiAlH4 with 1-dodecanethiol as a stabilizer and weak capping ligand.
As shown in Figure 1, at the end of the evaporation process the XRD pattern of the material confirmed the preparation of LiAlH4 with broader diffraction peaks indicating its nanostructure.The nanosized nature of the material was confirmed by TEM analysis showing isolated spherical nanoparticles with a particle size ranging from 2 to 16 nm (Figure 2).EDS analysis further confirmed the Al content of these nanoparticles that were thus believed to correspond to LiAlH4 (Figure 2f).In comparison, bulk LiAlH4 was imaged as large particles with no defined morphology (Figure 2a), hence the synthesis of nanosized LiAlH4 was successful.EDS analysis further confirmed the Al content of these nanoparticles that were thus believed to correspond to LiAlH 4 (Figure 2f).In comparison, bulk LiAlH 4 was imaged as large particles with no defined morphology (Figure 2a), hence the synthesis of nanosized LiAlH 4 was successful.
The hydrogen desorption properties of these nanoparticles were determined by TGA/DSC/MS.As shown in Figure 3, the as-prepared LiAlH 4 nanoparticles desorbed hydrogen in a similar way to bulk LiAlH 4 with the release of hydrogen occurring in two steps before 300 • C. The small additional exothermic peak at 145 • C was assigned to the reaction of the thiol head of the surfactant with LiAlH 4 at the surface of the nanoparticles.The similarities of the decompositions behavior is inherent to the inability of the surfactant to contain the melted LiAlH 4 and thus a loss of the nanosize feature in the favor of larger molten agglomerates behaving like bulk LiAlH 4 .The decomposition of the dodecanthiol as evidenced by mass spectrometry near 220 • C [35] further supports this hypothesis (Figure 3d).
The hydrogen desorption properties of these nanoparticles were determined by TGA/DSC/MS.As shown in Figure 3, the as-prepared LiAlH4 nanoparticles desorbed hydrogen in a similar way to bulk LiAlH4 with the release of hydrogen occurring in two steps before 300 °C.The small additional exothermic peak at 145 °C was assigned to the reaction of the thiol head of the surfactant with LiAlH4 at the surface of the nanoparticles.The similarities of the decompositions behavior is inherent to the inability of the surfactant to contain the melted LiAlH4 and thus a loss of the nanosize feature in the favor of larger molten agglomerates behaving like bulk LiAlH4.The decomposition of the dodecanthiol as evidenced by mass spectrometry near 220 °C [35] further supports this hypothesis (Figure 3d).and (c,d) respective hydrogen desorption profile as followed by mass spectrometry.The fragments C3H5 + and C3H7 + correspond to the decomposition of the 1-dodecanethiol surfactant [36].The decreasing heat flow peaks correspond to an endothermic process.
In order to better stabilize these nanoparticles and contain the melt, we aimed at a core-shell strategy [20].whereby the core LiAlH4 is contained within a metallic shell.To this aim, Ti was thought to be the best option because it is a well-known catalyst for other alanate systems including NaAlH4 and Mg(AlH4)2 [21,26,37].In particular, Ti-based additives have shown benefit for enhancing dehydrogenation kinetics and reduction of the decomposition temperature of alanates [9].Ti also forms a hydride and has good hydrogen diffusion properties [38,39].In addition, TiCl3 can be readily reduced by LiAlH4, and thus could form through the transmetalation method [19,20,31].A core-shell structure assuming that the reduction rate of the titanium precursor at the surface of the LiAlH4 nanoparticles can be controlled to form a continuous shell [40].
To coat the nanosized LiAlH4 particles with Ti, LiAlH4 nanoparticles were suspended in pentane where they are insoluble and a solution of TiCl3 in pentane was added dropwise.In this process, TiCl3 is reduced at the surface of LiAlH4 nanoparticles following reaction (1) 3LiAlH4 + TiCl3 → 3LiCl + Ti + 3Al + 6H2 (1) The decreasing heat flow peaks correspond to an endothermic process.
In order to better stabilize these nanoparticles and contain the melt, we aimed at a core-shell strategy [20].whereby the core LiAlH 4 is contained within a metallic shell.To this aim, Ti was thought to be the best option because it is a well-known catalyst for other alanate systems including NaAlH 4 and Mg(AlH 4 ) 2 [21,26,37].In particular, Ti-based additives have shown benefit for enhancing dehydrogenation kinetics and reduction of the decomposition temperature of alanates [9].Ti also forms a hydride and has good hydrogen diffusion properties [38,39].In addition, TiCl 3 can be readily reduced by LiAlH 4 , and thus could form through the transmetalation method [19,20,31].A core-shell structure assuming that the reduction rate of the titanium precursor at the surface of the LiAlH 4 nanoparticles can be controlled to form a continuous shell [40].
To coat the nanosized LiAlH 4 particles with Ti, LiAlH 4 nanoparticles were suspended in pentane where they are insoluble and a solution of TiCl 3 in pentane was added dropwise.In this process, TiCl 3 is reduced at the surface of LiAlH 4 nanoparticles following reaction (1).
After ageing overnight, washing, and drying, the resulting black suspension was characterized.As shown by XRD, the diffraction pattern of the material mainly contained peaks with a weaker intensity related to the monoclinic phase of LiAlH 4 and additional peaks assigned to tetragonal TiH and residual TiCl 3 (Figure 4).No metallic Ti was observed and this was consistent with the ability of LiAlH 4 to reduce metal halides with the formation of their corresponding hydride [41].Non-stoichiometric TiH 2−x has previously been reported to form when heating metallic Ti under hydrogen pressure at temperatures >300 • C [39,42].However, under the current experimental conditions, it is possible that the hydrogen release upon reaction (1) directly interacts with the newly formed Ti nuclei to generate the corresponding hydride.The remaining TiCl 3 observed by XRD was unexpected owing to the large molar ratio (16:1) of LiAlH 4 compared to TiCl 3 , and this indicated that the reduction of TiCl 3 by LiAlH 4 stopped during the process.This could be the result of the formation of a TiH layer at the surface of the LiAlH 4 nanoparticles significantly decreasing the contact area of LiAlH 4 and TiCl 3 , and thus precluding any further reduction.
Inorganics 2017, 5, 38 5 of 11 After ageing overnight, washing, and drying, the resulting black suspension was characterized.As shown by XRD, the diffraction pattern of the material mainly contained peaks with a weaker intensity related to the monoclinic phase of LiAlH4 and additional peaks assigned to tetragonal TiH and residual TiCl3 (Figure 4).No metallic Ti was observed and this was consistent with the ability of LiAlH4 to reduce metal halides with the formation of their corresponding hydride [41].Non-stoichiometric TiH2−x has previously been reported to form when heating metallic Ti under hydrogen pressure at temperatures >300 °C [39,42].However, under the current experimental conditions, it is possible that the hydrogen release upon reaction (1) directly interacts with the newly formed Ti nuclei to generate the corresponding hydride.The remaining TiCl3 observed by XRD was unexpected owing to the large molar ratio (16:1) of LiAlH4 compared to TiCl3, and this indicated that the reduction of TiCl3 by LiAlH4 stopped during the process.This could be the result of the formation of a TiH layer at the surface of the LiAlH4 nanoparticles significantly decreasing the contact area of LiAlH4 and TiCl3, and thus precluding any further reduction.Additional characterization by TEM revealed largely spherical particles with a broad particle size distribution ranging from 5 to 50 nm (Figure 5a,b,e).This corresponds to a significant increase in particle size as compared to the uncoated LiAlH4 nanoparticles (Figure 2) and thus indicates the deposition of a TiH layer.Indeed, EDS analysis and elemental mapping revealed the Ti and Al content of the nanoparticles imaged and the even distribution of Ti on the LiAlH4 nanoparticles (Figure 5c,d).Hence, it was believed that Ti was successfully deposited at the surface of the LiAH4 nanoparticles.The weaker intensity of the XRD peaks related to LiAlH4 in the Ti coated nanosized LiAlH4 material also corroborates this conclusion (Figure 4), because upon coating the diffraction peak intensity of the core has been reported to significantly decrease [31].
Analysis of the hydrogen desorption properties of these Ti modified nanoparticles revealed a unique behavior (Figure 6).Nanosized LiAlH4 modified with Ti almost fully decomposed through a single and extremely exothermic event recorded by DSC at 120 °C (Figure 6a,b).In comparison, bulk LiAlH4 ball milled with TiCl3 (material denoted bulk-LiAlH4/Ti) decomposed following the known decomposition path of LiAlH4-i.e., via Steps 1 and 2 (Figure 6c,d)-in agreement with previous reports [9,43,44].This was confirmed by the XRD of bulk LiAlH4/Ti after heating at 150, 200 and 300 °C.
As shown in Figure 6b, at 150 °C, bulk-LiAlH4/Ti started partial decomposition to Li3AlH6 according to Step 1. Then at 180 °C, the LiAlH4 phase totally converted to Li3AlH6 and at 300 °C, Li3AlH6 fully decomposed to LiH and Al following Step 2. In comparison, the diffraction pattern of nanosized LiAlH4 coated with Ti heated at 150 °C after the single exothermic event showed that no LiAlH4 remained, while only a very small amount of Li3AlH6 was detected (Figure 7a).The Additional characterization by TEM revealed largely spherical particles with a broad particle size distribution ranging from 5 to 50 nm (Figure 5a,b,e).This corresponds to a significant increase in particle size as compared to the uncoated LiAlH 4 nanoparticles (Figure 2) and thus indicates the deposition of a TiH layer.Indeed, EDS analysis and elemental mapping revealed the Ti and Al content of the nanoparticles imaged and the even distribution of Ti on the LiAlH 4 nanoparticles (Figure 5c,d).Hence, it was believed that Ti was successfully deposited at the surface of the LiAH 4 nanoparticles.The weaker intensity of the XRD peaks related to LiAlH 4 in the Ti coated nanosized LiAlH 4 material also corroborates this conclusion (Figure 4), because upon coating the diffraction peak intensity of the core has been reported to significantly decrease [31].
Analysis of the hydrogen desorption properties of these Ti modified nanoparticles revealed a unique behavior (Figure 6).Nanosized LiAlH 4 modified with Ti almost fully decomposed through a single and extremely exothermic event recorded by DSC at 120 • C (Figure 6a  C after the single exothermic event showed that no LiAlH 4 remained, while only a very small amount of Li 3 AlH 6 was detected (Figure 7a).The remaining Li 3 AlH 6 could be the result of unevenly or partially coated LiAlH 4 nanoparticles evolving to bulk size materials upon melting.However, this could not be confirmed since no apparent melting was detected by DSC peak (Figure 6a).remaining Li3AlH6 could be the result of unevenly or partially coated LiAlH4 nanoparticles evolving to bulk size materials upon melting.However, this could not be confirmed since no apparent melting was detected by DSC peak (Figure 6a).remaining Li3AlH6 could be the result of unevenly or partially coated LiAlH4 nanoparticles evolving to bulk size materials upon melting.However, this could not be confirmed since no apparent melting was detected by DSC peak (Figure 6a).Another possibility for the partial decomposition of Li3AlH6 is that the intended core-hell structure disintegrated under the sharp exothermic event as proven by TEM analysis of the dehydrogenated material (Figure 8), and consequently sprayed around undecomposed Li3AlH6.The observed suppression of melting as per bulk LiAlH4 is also an indicator that LiAlH4 was confined within a TiH shell.The possible core-shell structure disintegration could also be the reason of an appearance of LiCl and the disappearance of diffraction peaks related to TiH (Figure 7a).LiCl was not present in the as-prepared Ti coated nanosized LiAlH4 (Figure 4), and this potentially indicates that more LiAlH4 was exposed to react with TiCl3 to form LiCl following reaction (1) during the exothermic event.The lack of TiH phase after the exothermic release of hydrogen also indicates a deterioration of the coating owing to the heat generated during the exothermic release of hydrogen.TiH has been reported to decompose at temperatures above 470 °C [45].
Furthermore, upon hydrogen release, nanosized LiAlH4 coated with Ti lost about 5.7 mass % almost instantly near 120 °C.Taking into account the weight of TiCl3 added to nanosized LiAlH4, the theoretical hydrogen release for Step 1 should be reduced from 5.3 to 4.2 mass %, and from 2.6 to 2.1 mass % for both Steps 2 and 3.This corresponds to a theoretical hydrogen content of 6.3 mass % up to Step 2. The 5.7 mass % loss of the Ti coated nanosized LiAlH4 thus corresponds to a close to full hydrogen release.Possible hydrogen release from TiH was unlikely to be noticeable in the hydrogen desorption profile.Assuming that nanoscale TiH was decomposed, this would correspond to an additional amount of hydrogen of less than 0.1 mass %.Accordingly, this result implies a significant thermodynamic alternation induced by the synergetic effects of the Ti catalytic Another possibility for the partial decomposition of Li 3 AlH 6 is that the intended core-hell structure disintegrated under the sharp exothermic event as proven by TEM analysis of the dehydrogenated material (Figure 8), and consequently sprayed around undecomposed Li 3 AlH 6 .The observed suppression of melting as per bulk LiAlH 4 is also an indicator that LiAlH 4 was confined within a TiH shell.Another possibility for the partial decomposition of Li3AlH6 is that the intended core-hell structure disintegrated under the sharp exothermic event as proven by TEM analysis of the dehydrogenated material (Figure 8), and consequently sprayed around undecomposed Li3AlH6.The observed suppression of melting as per bulk LiAlH4 is also an indicator that LiAlH4 was confined within a TiH shell.The possible core-shell structure disintegration could also be the reason of an appearance of LiCl and the disappearance of diffraction peaks related to TiH (Figure 7a).LiCl was not present in the as-prepared Ti coated nanosized LiAlH4 (Figure 4), and this potentially indicates that more LiAlH4 was exposed to react with TiCl3 to form LiCl following reaction (1) during the exothermic event.The lack of TiH phase after the exothermic release of hydrogen also indicates a deterioration of the coating owing to the heat generated during the exothermic release of hydrogen.TiH has been reported to decompose at temperatures above 470 °C [45].
Furthermore, upon hydrogen release, nanosized LiAlH4 coated with Ti lost about 5.7 mass % almost instantly near 120 °C.Taking into account the weight of TiCl3 added to nanosized LiAlH4, the theoretical hydrogen release for Step 1 should be reduced from 5.3 to 4.2 mass %, and from 2.6 to 2.1 mass % for both Steps 2 and 3.This corresponds to a theoretical hydrogen content of 6.3 mass % up to Step 2. The 5.7 mass % loss of the Ti coated nanosized LiAlH4 thus corresponds to a close to full hydrogen release.Possible hydrogen release from TiH was unlikely to be noticeable in the hydrogen desorption profile.Assuming that nanoscale TiH was decomposed, this would correspond to an additional amount of hydrogen of less than 0.1 mass %.Accordingly, this result implies a significant thermodynamic alternation induced by the synergetic effects of the Ti catalytic The possible core-shell structure disintegration could also be the reason of an appearance of LiCl and the disappearance of diffraction peaks related to TiH (Figure 7a).LiCl was not present in the as-prepared Ti coated nanosized LiAlH 4 (Figure 4), and this potentially indicates that more LiAlH 4 was exposed to react with TiCl 3 to form LiCl following reaction (1) during the exothermic event.The lack of TiH phase after the exothermic release of hydrogen also indicates a deterioration of the coating owing to the heat generated during the exothermic release of hydrogen.TiH has been reported to decompose at temperatures above 470 • C [45].
Furthermore, upon hydrogen release, nanosized LiAlH 4 coated with Ti lost about 5.7 mass % almost instantly near 120 • C. Taking into account the weight of TiCl 3 added to nanosized LiAlH 4 , the theoretical hydrogen release for Step 1 should be reduced from 5.3 to 4.2 mass %, and from 2.6 to 2.1 mass % for both Steps 2 and 3.This corresponds to a theoretical hydrogen content of 6.3 mass % up to Step 2. The 5.7 mass % loss of the Ti coated nanosized LiAlH 4 thus corresponds to a close to full hydrogen release.Possible hydrogen release from TiH was unlikely to be noticeable in the hydrogen desorption profile.Assuming that nanoscale TiH was decomposed, this would correspond to an additional amount of hydrogen of less than 0.1 mass %.Accordingly, this result implies a significant thermodynamic alternation induced by the synergetic effects of the Ti catalytic coating with LiAlH 4 nanosizing potentially involving the decompositions of both LiAlH 4 and Li 3 AlH 6 in a rapid manner following the direct reaction (2).
Although, the currently modified properties of Ti modified nanosized LiAHl 4 are unsuitable for hydrogen storage purposes, this finding implies that the strategy of catalytic coating on hydride nanoparticles offers a more versatile avenue to modifying the hydrogen storage properties of a hydride to a larger extent when compared to mechanically activated LiAlH 4 with Ti.

Materials and Methods
All experiments were performed under an inert atmosphere in an argon filled glove box (O 2 and H 2 O < 1 ppm) from LC Technology (Salisbury, MD, USA) LiAlH 4 2.0 M in THF, titanium (III) chloride, 1-dodecanethiol, and anhydrous pentane were purchased from Sigma Aldrich (Sydney, Australia) and used as received.Lithium aluminum hydride (LiAlH 4 , 95%) was purchased from Sigma Aldrich (Sydney, Australia), and purified by dissolving it in a large amount of diethyl ether and recrystallizing it from the filtrate solution.

Synthesis of LiAlH 4 Nanoparticles
In a typical synthesis, 5 mL of a commercial LiAlH 4 solution (2.0 M in THF) was placed in a vial and stirred gently with 10 µL (2 mg) of 1-dodecanethiol.The vial was then closed tightly and the solution was left to age overnight.The solvent was then evaporated under a moderate vacuum until a viscous paste was formed.The later was fully dried at 30 • C under dynamic vacuum.The yield was around 410 mg, which is slightly higher than the 382 mg theoretical yield owing to remaining THF bonded to LiAlH 4.

Coating of the LiAlH 4 Nanoparticles with Ti
The as-synthesized LiAlH 4 nanoparticles (100 mg) were suspended in 20 mL of pentane, under magnetic stirring at 600 rpm to form a milky suspension.25 mg of TiCl 3 (8.1 10 −3 mol•L −1 ) was then suspended in a 10 mL solution of pentane and then added dropwise to the suspension of the LiAlH 4 nanoparticles at a rate of 50 µL•min −1 .After a few minutes, the suspension turned black and was allowed to age overnight before separation by centrifugation and drying under dynamic vacuum overnight.

Preparation of Reference Material Ball Milled LiAlH 4 with TiCl 3
As a reference, commercial bulk LiAlH 4 was ball milled with TiCl 3 (5 mol %) using a Retsch Mixer Mill MM 400 (Haan, Germany).The mixture was milled for 10 min at 20 Hz three times with a ball to powder weight ratio of 135:1.

Characterization
Transmission Electron Microscopy (TEM) and Energy Dispersive X-ray Spectroscopy (EDS) were performed with a Philips CM200 (Sydney, Australia) operated at 200 kV.The materials were dispersed in pentane, dropped onto a carbon coated copper grid, and dried in an argon filled glovebox before transfer to the microscope in a quick manner as to minimize air exposure.X-ray Diffraction (XRD) was performed by using a PANalytical X'pert Multipurpose XRD system (Sydney, Australia) operated at 40 mA and 45 kV with a monochromated Cu Kα radiation (λ = 1.541Å); step size = 0.01, 0.02 or 0.05; time per step = 10 or 20 s/step.The materials were protected against oxidation from air by a Kapton foil covering a stainless steel sample holder.Hydrogen desorption profiles were acquired by Thermogravimetric Analysis (TGA)/Differential Scanning Calorimetry (DSC) coupled with Mass Spectrometry (MS) using a Mettler Toledo TGA/DSC (Sydney, Australia) 1 coupled with an Omnistar MS.Measurements were conducted with alumina crucibles at 10 • C•min −1 under an argon flow of 25 mL•min −1 .Masses between m/z = 2 and 100 were recorded.Reversibility was checked by manually performing hydrogen absorption/desorption cycles with a homemade Sievert apparatus and a stainless steel sample holder operated at 300 • C, with 7 MPa hydrogen pressure for absorption and 0.01 MPa for desorption.No significant reversibility was observed.

Conclusions
An effective downsizing and coating approach was developed and demonstrated that it is possible to synthesize isolated LiAlH 4 nanoparticles to a certain extent and ultimately modify the hydrogen storage properties of LiAlH 4 .Once coated with Ti, nanosized LiAlH 4 was found to decompose at 120 • C in a single exothermic event with a 5.7 mass % and instant hydrogen release.This finding implies a significant thermodynamic alternation induced by the synergetic effects of the Ti catalytic coating with LiAlH 4 nanosizing that ultimately changed its decomposition path to LiAlH 4 → Al + LiH + 3/2H 2 .This finding also demonstrates that the hydrogen storage properties of LiAlH 4 should be size dependent and thus provides a new avenue to modify the properties of LiAlH 4 through particle size effects.Due to the violent exothermic nature of its decomposition, this material is considered not viable for practical hydrogen storage purposes.However, Ti coated nanosized LiAlH 4 offers a unique alternative as compared to hydrides which only supply hydrogen via an endothermic reaction.It enables applications where larger amounts of hydrogen and heat are both needed at the same time and rapidly, e.g., for compact explosives and fire starters.

Figure 2 .
Figure 2. TEM images of (a) commercial bulk LiAlH4 and (b-d) as-prepared nanosized LiAlH4 via solvent evaporation; (e) corresponding particle size distribution and (f) EDS spectrum.

Figure 2 .
Figure 2. TEM images of (a) commercial bulk LiAlH4 and (b-d) as-prepared nanosized LiAlH4 via solvent evaporation; (e) corresponding particle size distribution and (f) EDS spectrum.

Figure 3 .
Figure 3. TGA/DSC for (a) commercial bulk LiAlH4; (b) as-prepared nanosized LiAlH4 via solvent evaporation,and (c,d) respective hydrogen desorption profile as followed by mass spectrometry.The fragments C3H5 + and C3H7 + correspond to the decomposition of the 1-dodecanethiol surfactant[36].The decreasing heat flow peaks correspond to an endothermic process.

Figure 5 .
Figure 5. TEM images of (a,b) as-prepared Ti coated nanosized LiAlH4 and corresponding (c) elemental mapping analysis; (d) EDS spectrum and (e) particle size distribution.

Figure 6 .
Figure 6.TGA/DSC for (a) as-prepared Ti coated nanosized LiAlH4, (c) commercial bulk LiAlH4 ball milled with TiCl3, (b,d) respective hydrogen desorption profile as followed by mass spectrometry.No other gases than hydrogen were detected.

Figure 5 .
Figure 5. TEM images of (a,b) as-prepared Ti coated nanosized LiAlH 4 and corresponding (c) elemental mapping analysis; (d) EDS spectrum and (e) particle size distribution.

Figure 5 .
Figure 5. TEM images of (a,b) as-prepared Ti coated nanosized LiAlH4 and corresponding (c) elemental mapping analysis; (d) EDS spectrum and (e) particle size distribution.

Figure 6 .
Figure 6.TGA/DSC for (a) as-prepared Ti coated nanosized LiAlH4, (c) commercial bulk LiAlH4 ball milled with TiCl3, (b,d) respective hydrogen desorption profile as followed by mass spectrometry.No other gases than hydrogen were detected.

Figure 6 .
Figure 6.TGA/DSC for (a) as-prepared Ti coated nanosized LiAlH 4 , (c) commercial bulk LiAlH 4 ball milled with TiCl 3 , (b,d) respective hydrogen desorption profile as followed by mass spectrometry.No other gases than hydrogen were detected.

Figure 7 .
Figure 7. XRD of (a) as-prepared Ti coated nanosized LiAlH4 after hydrogen release at 150 °C and (b) commercial bulk LiAlH4 ball milled with TiCl3 after thermal decomposition at 150, 180 and 300 °C.

Figure 8 .
Figure 8. TEM images of (a) as-prepared Ti coated nanosized LiAlH4 after hydrogen release at 150 °C and (b) corresponding elemental mapping analysis.

Figure 8 .
Figure 8. TEM images of (a) as-prepared Ti coated nanosized LiAlH4 after hydrogen release at 150 °C and (b) corresponding elemental mapping analysis.

Figure 8 .
Figure 8. TEM images of (a) as-prepared Ti coated nanosized LiAlH 4 after hydrogen release at 150 • C and (b) corresponding elemental mapping analysis.