The influence of ball-milling time on the dehydrogenationproperties of the NaAlH4-MgH2 compositeCitation for published version (APA):Bendyna, J. K., Dyjak, S. M., & Notten, P. H. L. (2015). The influence of ball-milling time on the dehydrogenationproperties of the NaAlH4-MgH2 composite. International Journal of Hydrogen Energy, 40(11), 4200-4206.https://doi.org/10.1016/j.ijhydene.2015.01.026

DOI:10.1016/j.ijhydene.2015.01.026

Document status and date:Published: 01/01/2015

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:openaccess@tue.nlproviding details and we will investigate your claim.

Download date: 23. Feb. 2022

ww.sciencedirect.com

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 6

Available online at w

ScienceDirect

journal homepage: www.elsevier .com/locate/he

The influence of ball-milling time on thedehydrogenation properties of the NaAlH4eMgH2

composite

Joanna K. Bendyna a,b, Sławomir Dyjak c,*, Peter H.L. Notten a,d

a Eindhoven University of Technology, Department of Chemical Engineering and Chemistry, Den Dolech 2, Postbus

513, 5600 MB Eindhoven, The Netherlandsb EMPA Swiss Federal Laboratories for Materials Science and Technology, Dept. Energy, Environment and Mobility,

CH-8600 Dubendorf, Switzerlandc Military University of Technology, Institute of Chemistry, 2 Kaliskiego Street, 00-908 Warsaw, Polandd Forschungszentrum Julich IEK-9, D-52425 Julich, Germany

a r t i c l e i n f o

Article history:

Received 12 November 2014

Received in revised form

6 January 2015

Accepted 8 January 2015

Available online 21 February 2015

Keywords:

Metal hydrides

Dehydrogenation reaction pathways

Ball milling

Alloys

X-ray diffraction

* Corresponding author. Military University oWarsaw, Poland. Tel.: þ48 (22) 683 77 46.

E-mail addresses: bendyna@gmail.com (Jhttp://dx.doi.org/10.1016/j.ijhydene.2015.01.00360-3199/Copyright © 2015, Hydrogen Energ

a b s t r a c t

The recently developed NaAlH4eMgH2 composite shows improved hydrogen-storage

properties compared to MgH2 and NaAlH4. However, the dehydrogenation reaction rates

are still too limited, hampering practical applications. Mechanical ball milling is broadly

used to improve the dehydrogenation reaction rates of hydrides. Therefore, the hydrogen-

storage properties of the NaAlH4eMgH2 (1:1) composite have been investigated as a func-

tion of ball-milling time. Expectedly, elongated milling led to a faster dehydrogenation

rates. New insights of the structural transformation pathways of the decomposition re-

action are provided. A number of AleMg alloys, including the only reported Al12Mg17, seem

to participate in the dehydrogenation. Thereby, complex alloying process of Al with Mg or

MgH2 has been proposed. Our data indicate the possibility that hydrides Na3AlH6 and

NaMgH3, which are the intermediate products of the dehydrogenation, coexist. The study

shed a light on the complexity of the decomposition pathways of hydride mixtures in

which the key role play alloys.

Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

Introduction

Nowadays, the growing energy demand and increasing levels

of green-house gasses urges the development of alternative

energy solutions [1,2]. Amongst many potential solutions a

hydrogen-based economy attracts considerable attention.

This is based on the so-called ‘green energy’ concept that is

f Technology, Institute o

.K. Bendyna), sdyjak@wa26y Publications, LLC. Publ

realised as a closed cycle process. At the start of this cycle,

hydrogen can directly be produced from water by using sun-

light, e.g. by photo-catalysis or indirectly by combining solar

cells with electrolysis of water [3]. Subsequently, hydrogen

must be stored efficiently for a safe transport. Finally, the

green cycle is closed when hydrogen is oxidized together with

oxygen to deliver energy and water [4,5]. However,

f Chemistry, Department of Explosives, 2 Kaliskiego Street, 00-908

t.edu.pl (S. Dyjak), P.H.L.Notten@tue.nl (P.H.L. Notten).

ished by Elsevier Ltd. All rights reserved.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 6 4201

implementation of the hydrogen-based economy suffers of a

severe bottleneck. Efficient and safe hydrogen-storage is still

very challenging, especially for the automotive industry ap-

plications [6].

The most promising methods to store hydrogen are by

means of chemisorption and physisorption. In the latter case,

most attractive sorbents are used with a large surface area

that requires very low temperatures, as low as liquid nitrogen,

to storage the hydrogen. These methods reveal good sorption

kinetics and often reasonable storage capacities [7]. The

chemisorption method is mostly dominated by alkali- and

alkaline-earth metal hydrides that often require elevated

temperatures for sorption and show high storage capacity. In

this group a large number of compounds are composed of

light-weight metals like Li and Na.

Both metals are able to form so-called alanates, such as

LiAlH4 [8,9] and NaAlH4 [9]. These alanates have been inten-

sively investigated as prospective complex hydrides with a

high theoretical gravimetric storage capacity of 10.6 and

7.5 wt.% for LiAlH4 and NaAlH4, respectively [10].

Thermodynamic destabilization of complex hydrides can

be achieved by using an additional hydride that results in the

formation of intermediate phases. Attractive hydride com-

posites have been developed, like LiAlH4eMgH2 [11] and

NaAlH4eMgH2 [12e14], which have a high storage capacity of

9.4 and 7.6 wt.%, respectively. So far, much emphasis in the

hydrogen-storage community was dedicated to the

LiAlH4eMgH2 system. Regarding the NaAlH4eMgH2 system

only a few limited investigations have been carried out, which

leave much space for further improvements of the composite

and to get more in-depth knowledge.

The process of the thermodynamic destabilization of

LiAlH4eMgH2 and NaAlH4eMgH2 involves complex multi-step

reactions. Firstly, LiAlH4 and NaAlH4 are expected to form

intermediate complex hydrides (Li3AlH6 or Na3AlH6) during

the decomposition. Furthermore, hydrogen gas, Al and metal

hydrides (LiH or NaH) are formed, according to a two-step

reaction [11e14].

3Li=NaAlH4/Li=Na3AlH6 þ 3Alþ 3H2 (1)

Li=Na3AlH6/3Li=NaHþ Alþ 1:5H2 (2)

The studies in which various molar ratios of LiAlH4:MgH2

(1:1, 1:2 and 1:4) were used, revealed that the dehydrogena-

tion pathway follows always the reactions (1) and (2) [15]. It

has been described that Li0.92Mg4.08 and Al12Mg17 alloys are

formed as well [11,16], according to the following reactions

4.08 MgH2 þ 0.92 LiH / Li0.92Mg4.08 þ 4.5 H2 (3)

12 Al þ 17 MgH2 / Al12Mg17 þ 17 H2 (4)

Reaction (3) has been described in more general terms in

Ref. [17].

n AHx þ m BHy 4 AnBm þ 1/2(nx þ my) H2 (5)

It is believed that the above alloys play a dominant role in

the thermodynamic destabilization of the LiAlH4eMgH2

composites [11,15e17].

The investigation of NaAlH4eMgH2 [14] revealed two

possible reaction schemes (1:4 and 1:2 vs. 1:1), unlike

LiAlH4eMgH2 (1:1, 1:2 and 1:4) [11]. The NaAlH4eMgH2 (1:1)

composition [13,14] is expected to form the dehydrogenation

product Na3AlH6 and follow the anticipated reactions (1), (2)

and (4) [11,13,14]. Further decomposition of NaH and resid-

ual MgH2 leads to hydrogen gas and respectivemetals without

the formation of a NaeMg alloy.

The NaAlH4eMgH2 (1:4 and 1:2) composites are proposed to

dehydrogenate via intermediate products such as Al12Mg17(reaction (4)) and NaMgH3 [12e14], without the formation of

Na3AlH6. It is expected that the formation of the Al12Mg17 alloy

during dehydrogenation improves the thermodynamic prop-

erties of the NaAlH4eMgH2 system [12,14]. NaMgH3 is formed

according to

NaAlH4 þ MgH2 / NaMgH3 þ Al þ 1.5 H2 (6)

It is not clear yet why the reaction (6) was not observed in

the NaAlH4eMgH2 composites with the molar ratio of 1:1

[13,14] but only in the mixtures with excess MgH2 [12,14].

Furthermore, the decomposition of the excess MgH2 into Mg

and hydrogen as well as reaction (4) take place. At a later stage

of the dehydrogenation process, NaH and NaMgH3 decompose

according to

NaMgH3 / NaH þ Mg þ H2 (7)

NaH / Na þ 0.5 H2 (8)

Recently, it has been reported that during the dehydroge-

nation of the Na3AlH6eMgH2 mixture, NaMgH3 can also be

formed [18].

Na3AlH6 þ 3 MgH2 / 3 NaMgH3 þ Al þ 1.5 H2 (9)

The decomposition reactions of the NaAlH4eMgH2 and

Na3AlH6eMgH2 composites as presented above indicate the

possibility of a coexistence of the intermediate hydrides

Na3AlH6 and NaMgH3 since both can be formed from the same

hydride mixtures. So far, it was suggested that the reported

NaAlH4eMgH2 mixture dehydrogenates exclusively via one

intermediate hydride NaMgH3 or Na3AlH6 which depends on

the molar ratio of the primary hydrides. However, the X-ray

diffraction (XRD) reflections of these ball-milled samples may

show much overlap, since the crystallinity is typically signif-

icantly decreased by grinding, which may lead to issues with

the data interpretation.

The goals of the present study are twofold. Firstly, to

analyse the influence of the high-energy ball-milling on the

dehydrogenation properties as a function of time and sec-

ondly, to gain more knowledge about the structural trans-

formation pathways of the NaAlH4eMgH2 (1:1) composite.

Fig. 1 e The dehydrogenation characteristics of the NaAlH4eMgH2 (1:1) mixture ball-milled for 1e360 min with a high-

energy ball-mill: a) desorbed hydrogen as a function of the temperature and b) desorbed hydrogen and temperature as a

function of time.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 64202

Experimental methods

All samples were handled and manipulated in an argon filled

glove-box. Sodium Aluminium Hydride (NaAlH4, Sigma-

eAldrich, 95%) and Magnesium Hydride (MgH2, Sigma-

eAldrich, 95%) were used as-received from producer without

further purification.

2 g of powder mixtures of NaAlH4eMgH2 (1:1) were placed

in a stainless steel 50 ml vial and milled with a shaker SPEX

8000 mill which operates at 1200 rpm and subsequently,

placed in the glove-box. Stainless steel balls (6 of 13 mm and

16 of 6 mm diameter) were used for the milling experiments.

The ball-to-powder weight ratio was kept constantly at 16:1.

The samples were milled from 1 to 360 min.

The powder samples were characterized, before and after

milling, with X-ray diffraction (XRD) using a Bruker D4

Endeavor X-ray Diffractometer with Cu Ka radiation and 2q

range of 10e80� and 0.02� step. Thematerials were loaded in a

commercial steel sample holder equipped with a Kapton foil

to protect the sample fromair. The ICDDdata basewas used to

evaluate the phase formation.

Dehydrogenation experiments were conducted with a

Sievert-type apparatus HyEnergy PCTpro2000. The Sievert-

type of set-up was successfully applied before to investigate

temperature programmed desorption (TPD) [19]. The samples

were prepared and measured applying the standard proce-

dure for each step. This includes a leak check; purge gas and

volume calibration at 100 �C. The dehydrogenation experi-

ments were performed in a static vacuum (staring level

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 6 4203

~0.1 bar). The TPD measurements as conducted between 100

and 350 �C with a heating rate of 3 �C/min. After approxi-

mately 4 or 2 h the pressure was reverted to the starting level

due to the pressure increase caused by hydrogen release from

themixture and temperature change. Therefore, blank tests in

the static vacuum were performed under similar TPD condi-

tions to estimate the effect of the pressure change and effec-

tive wt.% changes of the performed experiments. The blank

tests have shown small wt.% changes of the order of

~0.01e0.05 wt.%.

Fig. 2 e XRD diffraction patterns collected at room

temperature of the NaAlH4eMgH2 mixtures after 1 min to

6 h of ball-milling: a) 2q range of 29e66� and b) 2q range of

29e41�.

Results

Dehydrogenation

The effect of ball-milling time on the dehydrogenation

behaviour of the NaAlH4eMgH2 (1:1) mixture is shown in

Fig. 1. Prolonged ball-milling (30e360min) has an effect on the

dehydrogenation reaction rates. For these samples the initial

temperature at which hydrogen starts to get released is 150 �C,which is approximately 30 �C lower compared to the sample

milled for only 1 min (Fig. 1a). This effect may be associated

with the impact of high energy ball-milling on the grinded

samples. Prolonged milling of hydrides such as MgH2 leads to

decreased crystallite and particle sizes, and of the introduc-

tion of various defects, which enable a faster hydrogen

diffusion process in the bulk of the ball-milled materials

[20,21].

A significant temperature decrease of 60e80 �C, dependingon the sample, is also more clearly observed during the

completed first dehydrogenation step in Fig. 1a and b. The first

step of the dehydrogenation is completed at 290, 230, 210 and

220 �C for the 1, 30, 120 and 360 min milled samples, respec-

tively. These data indicate that the first step was stretched

over a large temperature range (between 180 and 290 �C) in the

sample milled for 1 min, while other samples released

hydrogen much faster in a significantly shorter temperature

range (from 150 to 210e230 �C). Typically, this step is associ-

ated with the formation of intermediate hydrides (see

reactions (1) and (6) [11e14]).

Further stages of dehydrogenation can be observed in

Fig. 1b in which, at constant temperature (350 �C), all the

milled compositions show two more distinctive dehydroge-

nation processes. These represent stages of further phase

transformations of the composites. A detailed structural XRD

analysis will be discussed below.

Ball milling also shows an effect on the amount of

hydrogen released. The lowest amount of released hydrogen

was found to correspond to the sample which was milled

during 1 min. This sample showed a weight decrease of

~5.2 wt.% at themaximum temperature of 350 �C compared to

the samples, which were ball-milled for 30 (~5.8 wt.%),

120 (~5.8 wt.%) and 360 min (~5.6 wt.%). Obviously, the

hydrogen release flow of the samplemilled for 1 min seems to

be quite unsteady and not continuous. This may be attributed

to bad homogeneity of the mixture. In contrast, the samples

milled for longer times show smooth hydrogen release curves.

Note that both samples ball-milled for 120 and 360 min show

similar dehydrogenation reaction rates and storage capacities,

therefore the optimummilling time in these experiments was

120 min since extended ball milling was not beneficial for the

rates and consumes more electricity.

Structural analysis

Fig. 2 shows the XRD results of the NaAlH4eMgH2 (1:1) mix-

tures ball-milled for 1e360min in a SPEXmill. Themain effect

of the prolongedmilling is the decreasing size of particles and

crystallites. This resulted in significant broadening of all re-

flections and vanishing of the smallest peaks of the starting

materials. The main starting materials NaAlH4, MgH2 and Mg

(always present as traces in the commercial magnesium hy-

dride) could be confirmed in the mixture milled for 1 min. In

the samples milled elongated times the traces of Mg vanished

in XRD due to, firstly, decreasing crystallites size and, sec-

ondly, potential reactions of themetal with other components

of the mixture.

Furthermore, several additional very small and broad XRD

peaks could be detected in the samples milled elongated

times. These peaks could be assigned to two phases which

typically are observed during the dehydrogenation

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 64204

(decomposition) of the NaAlH4eMgH2 mixture. The first group

of the reflections might be attributed to b-Na3AlH6 or NaMgH3

(marked with &) while the second group of reflections may be

fitted to Al, alternatively to Al0.95Mg0.05 (marked with x). Thedecomposition reactions may be sometimes initiated during

high energy ball-milling of various complex hydridemixtures.

The XRD of the dehydrogenated NaAlH4eMgH2 mixtures

which were previously ball-milled for 1 and 120 min are

shown in Fig. 3. For simplicity, only these two XRD patterns

are shown because the phase composition of the 120 min

sample resembles the two remaining samples. It should be

noted that the overall phase composition of all samples seems

to be somewhat similar, and the main differences are related

to the peak intensity of the components.

In these multi-component samples, the most distinct

phases that only partially overlap with other compounds are

NaH and Al12Mg17. A clear distinction between b-Na3AlH6 and

NaMgH3 is again hindered since the identification is exclu-

sively based on a few strongest reflections (many smallest are

missing) which are broad and overlap with each other and

some other phases.

Next to the above mentioned Al12Mg17 alloy, at least two

other alloys can be identified, such as AlMg and Al3Mg2. Also

other alloy candidates, assigned using some low intensity

reflections (not indicated on the figure for simplicity), may be

Fig. 3 e XRD diffraction patterns collected at room

temperature of the NaAlH4eMgH2 mixtures ball-milled (for

1 min and 2 h) and after dehydrogenation at 350 �C: a) 2qrange of 29e66� and b) 2q range of 31e41�.

proposed, such as Al5.15Mg3.15, Al3.16Mg1.84, and Al2Mg. How-

ever, most reflections of these phases overlap with other

proposed compounds. Many reflections of the smallest in-

tensity belonging to various AleMg alloys are observedmostly

as some disturbance in the background (very small and broad

peaks), therefore these are not shown since their assignment

is hampered by a limited interpretation.

Among the expected single metal compounds, such as Mg,

Al, and Na, none of these could be observed.

Discussion

Proposed dehydrogenation pathway for ball-milledNaAlH4eMgH2 (1:1)

Based on the above presented data we would like to propose

the following possible decomposition pathway of the

NaAlH4eMgH2 (1:1) mixture, Phase 1, as shown in Fig. 4. The

evaluation of the XRD data leads to the conclusion that

decomposition of themixture is already initiated during high-

energy ball-milling for 30 min and longer times. This is

possible because the mechanical grinding results in elevated

pressures and temperatures at the impact point of balls.

However, the major decomposition of the composite takes

place during TPD above 150 �C (Phase 2 below). For the 1 min

ball-milled sample, it is expected that our proposed dehy-

drogenation pathway is only initiated at elevated tempera-

tures during the TPD experiment as could be judged from XRD

(Phase 1 and Phase 2 occurring only during TPD). It is worth-

while to note that only a small quantity of hydrogen was

released during milling for 30 min and longer times, as can be

judged from the dehydrogenation experiments.

The first major step in the decomposition process of the

NaAlH4eMgH2 (1:1) composite may be ascribed to the forma-

tion of the intermediate hydride b-Na3AlH6, alternatively

NaMgH3, or both simultaneously. Literature indicates one

more possibility, reaction (9) [18], in which Na3AlH6 react with

MgH2 to form NaMgH3. According to the literature, the two

hydrides b-Na3AlH6 and NaMgH3 [12e14] are formed in a

similar temperature range as the observed during our TPD

study (first step of the dehydrogenation process). The evalu-

ation of the XRD data, even if challenging, allowed concluding

that coexistence of both hydrides is possible. Therefore, it

cannot be excluded that these hydrides form as a result of

competing reactions.

Fig. 4 e Possible decomposition pathway of NaAlH4eMgH2

(1:1) initiated by high-energy ball-milling e Phase 1.

Fig. 5 e The potential decomposition pathway of the ball-

milled NaAlH4eMgH2 (1:1) mixture observed at high-

temperature TPD experiments e Phase 2.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 6 4205

Additionally, Al and Al0.95Mg0.05 (Al1�xMgx) are the dehy-

drogenation products of the composite on the above shown

Fig. 4. The general term Al1�xMgx relates exclusively to the

molar ratio of Al:Mg, but not to the crystal structures of the

observed alloys. The Al metal is detected when b-Na3AlH6 and

NaMgH3 are formed, while Al0.95Mg0.05 may be formed from Al

and Mg (present in the starting hydrides) or Al and MgH2.

Proposed dehydrogenation pathway for NaAlH4eMgH2

(1:1) at high temperatures

At high-temperatures (TPD experiments) further decomposi-

tion of the ball-milled NaAlH4eMgH2 (1:1) mixture occurs,

Phase 2, as indicated in Fig. 5. In particular, the decomposition

of the b-Na3AlH6 and NaMgH3 hydrides is taking place and

several AleMg alloys (described by a general term Al1-xMgx)

can be formed. Previous research only identified Al12Mg17 as

being responsible for the improvements of the NaAlH4eMgH2

composite thermodynamics (independently of themolar ratio

of NaAlH4:MgH2) [12e14].

We believe that the alloying process may take place at all

steps of the decomposition (Phase 1 and Phase 2), as proposed

in Fig. 6. It starts with Al that may be doped with small amount

of Mg (Al0.95Mg0.05). Further, the reaction continues at high

temperatures with the formation of the aluminum rich alloys,

such as Al3Mg2, Al5.15Mg3.15, and Al2Mg. Subsequently, the alloy

with a unity molar ratio of Al:Mg (1:1) might be formed. Finally,

the Mg-rich alloy, previously reported as Al12Mg17, could be

identified. Therefore, the simplified alloy formula Al1�xMgx is

Fig. 6 e Proposed alloying process based on the

dehydrogenation experiments of NaAlH4eMgH2 (1:1).

used with x ¼ 0e0.7. With x ¼ 0, Al1�xMgx is a pure Al metal,

while Al1�xMgx with x ¼ 0.7 relates to Al12Mg17.

We believe that the discrepancies between literature and

what was observed in this study regarding the formation of

the alloys may be attributed to the different temperature

ranges used in the TPD studies, reaction times, and challenges

to satisfactory identify the decomposition products in XRD

data due to multi-phase samples and often poor crystalline

materials. Our experiments end at approximately 350 �Cbecause of practical reasons and this therefore results in the

formation of multiple alloys, while experiments from the

literature mostly end at much higher temperatures between

450 �C and 500 �C, and result in the formation of only a single

alloy with the composition: Al12Mg17 [12e14,18]. The latter

experiments also lead to the complete decomposition of NaH,

to the Na metal and hydrogen gas, which was not observed in

this study.

The remaining questions, that can be derived based on the

above discussion, are how these various AleMg alloys form

exactly and how these influence the reversibility and ther-

modynamic stability of the NaAlH4eMgH2 composite. These

issues are the subject of our further study that is currently

underway. An attempt to evaluate the participation of the two

hydrides, b-Na3AlH6 and NaMgH3, in the dehydrogenation

process will be a subject of a detailed study using IR spec-

troscopy. Our further investigations include the studies on the

effect of the catalysts on the properties of the mixture.

Conclusions

NaAlH4eMgH2 is an attractive material for future hydrogen-

storage applications. The influence of high-energy ball-mill-

ing, as a function of time, on the structural phase trans-

formations and on the dehydrogenation properties of

NaAlH4eMgH2 (1:1) was investigated. Expectedly, the pro-

longed ball-milling of the mixtures leads to significantly

enhanced dehydrogenation reaction rates, i.e. desorption of

hydrogen atmuch lower temperatures. Novel insights into the

decomposition reactions of the material were provided. A

number of AleMg alloys seem to participate in the dehydro-

genation via a complex alloying process of Al withMg orMgH2.

The obtained results suggest the two hydrides Na3AlH6 and

NaMgH3 coexist during decomposition of the mixture. The

study shed a light on the complexity of the decomposition

pathways of hydride mixtures in which the key role play

alloys.

Acknowledgements

The authors thank the Eindhoven University of Technology

and Department of Chemical Engineering and Chemistry;

EMPA Swiss Federal Laboratories for Materials Science and

Technology and Empa International Postdoctoral Fellowship

programme funded by FP7: People Marie-Curie action

COFUND (PCOFUND-GA-2010-267161) for financial support of

the research. Many thanks to Dr. A. Remhof, Dr. M. Polanski

and M.Sc. R. Kregting for critical reading of the manuscript.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 4 0 ( 2 0 1 5 ) 4 2 0 0e4 2 0 64206

r e f e r e n c e s

[1] Manivasagam TG, Kiraz K, Notten PHL. Electrochemical andoptical properties of magnesium-alloy hydrides reviewed.Crystals 2012;2:1410e33.

[2] Ritter JA, Ebner AD, Wang J, Zidan R. Implementing ahydrogen economy. Mater Today 2003;6:18e23.

[3] Holladay JD, Hu J, King DL, Wang Y. An overview ofhydrogen production technologies. Catal Today2009;139:244e60.

[4] Schlapbach L, Zuttel A. Hydrogen-storage materials formobile applications. Nature 2001;414:353e8.

[5] Nicoleti G. The hydrogen option for energy: a review oftechnical, environmental and economic aspects. Int JHydrogen Eng 1995;20:759e65.

[6] Das LM. On-board hydrogen storage system for automotiveapplication. Int J Hydrogen Eng 1996;21:789e800.

[7] Yurum Y, Taralp A, Veziroglu TN. Storage of hydrogen innanostructured carbon materials. Int J Hydrogen Eng2009;34:3784e98.

[8] Liu X, Langmi HW, Beattie SD, Azenwi FF, McGrady GS,Jensen CM. Ti-doped LiAlH4 for hydrogen storage: synthesis,catalyst loading and cycling performance. J Am Chem Soc2011;133:15593e7.

[9] Wan C, Ju X, Qi Y, Zhang Y, Wang S, Liu X, et al. SynchrotronXRD and XANES studies of cerium-doped NaAlH4:elucidation of doping induced structure changes andelectronic state. J Alloy Compd 2009;481:60e4.

[10] Grochala W, Edwards PP. Thermal decomposition of the non-interstitial hydrides for the storage and production ofhydrogen. Chem Rev 2004;104:1283e315.

[11] Zhang Y, Tian QF, Liu SS, Sun LX. The destabilizationmechanism and de/re-hydrogenation kinetics ofMgH2eLiAlH4 hydrogen storage system. J Power Sources2008;185:1514e8.

[12] Ismail M, Zhao Y, Yu XB, Dou SX. Improved hydrogen storageperformance of MgH2eNaAlH4 composite by addition of TiF3.Int J Hydrogen Eng 2012;37:8395e401.

[13] Ismail M, Zhao Y, Yu XB, Mao JF, Dou SX. The hydrogenstorage properties and reaction mechanism of theMgH2eNaAlH4 composite system. Int J Hydrogen Eng2011;36:9045e50.

[14] Rafi-ud-din, Xuanhui Q, Ping L, Zhang L, Ahmad M, Iqbal MZ,et al. Enhanced hydrogen storage performance forMgH2eNaAlH4 systemdthe effects of stoichiometry andNb2O5 nanoparticles on cycling behaviour. RSC Adv2012;2:4891e3.

[15] Varin RA, Czujko T, Wronski ZS. Nanomaterials for solidstate hydrogen storage. New York, NY: SpringerScienceþBusiness Media; 2009.

[16] Chen R, Wang X, Xu L, Chen L, Li S, Chen C. An investigationon the reaction mechanism of LiAlH4eMgH2 hydrogenstorage system. Mater Chem Phys 2010;124:83e7.

[17] Vajo JJ, Mertens F, Ahn CC, Bowman RC, Fultz B. Alteringhydrogen storage properties by hydride destabilizationthrough alloy formation: LiH and MgH2 destabilized with Si. JPhys Chem B 2004;108:13977e83.

[18] Ismail M, Zhao Y, Dou SX. An investigation on the hydrogenstorage properties and reaction mechanism of thedestabilized MgH2eNa3AlH6 (4:1) system. Int J Hydrogen Eng2013;38:1478e83.

[19] Liang C, Liu Y, Luo K, Li B, Gao M, Pan H, et al. Reactionpathways determined by mechanical milling process fordehydrogenation/hydrogenation of the LiNH2/MgH2 system.Chem Eur J 2010;16:693e702.

[20] Lang J, Huot J. A new approach to the processing of metalhydrides. J Alloy Compd 2011;509:L18e22.

[21] Paik B, Walton A, Mann V, Book D, Jones IP, Harris IR.Microstructure of ball milled MgH2 powders upon hydrogencycling: an electron microscopy study. Int J Hydrogen Eng2010;35:9012e20.