Synergistic hydrogen desorption behavior of magnesium aluminum hydride synthesized by mechano-chemical activation method

Article abstract of DOI:10.1016/j.jallcom.2012.02.066

A mechano-chemical activation synthesis (MCAS) is employed to fabricate Mg(AlH₄)₂ via milling the precursors, specifically NaAlH₄ and MgCl₂. The corresponding dehydrogenation behavior of the synthesized powders is investigated. The experimental results showed that incomplete synthesis or premature dehydrogenation may occur if the milling process was not properly controlled. The hydrogen content of each synthesized powder is determined by using a thermal gravimetric analyzer (TGA). The dehydrogenation reactions of the synthesized powders are investigated by employing ex situ X-ray diffraction (XRD), in situ synchrotron XRD and differential thermal analysis (DTA). The results showed that the incompletely synthesized powder consisted of residual NaAlH₄ in the synthesized Mg(AlH₄)₂, which demonstrated an initial dehydrogenation temperature as low as 100 °C and accompanied with a maximum amount (3.1 wt%) of H₂ released below 350 °C. The mutual catalytic effect of both NaAlH₄ and Mg(AlH₄)₂ on lowering their initial dehydrogenation temperature is confirmed.

Full text of DOI:10.1016/j.jallcom.2012.02.066

Journal of Alloys and Compounds 525 (2012) 126–132
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Synergistic hydrogen desorption behavior of magnesium aluminum hydride
synthesized by mechano-chemical activation method
Cheng-Hsien Yang, Tzu-Teng Chen, Wen-Ta Tsai∗
Department of Materials Science and Engineering, National Cheng Kung University, Tainan 701, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
A mechano-chemical activation synthesis (MCAS) is employed to fabricate Mg(AlH4)2 via milling the
precursors, specifically NaAlH4 and MgCl2. The corresponding dehydrogenation behavior of the synthe-
sized powders is investigated. The experimental results showed that incomplete synthesis or premature
dehydrogenation may occur if the milling process was not properly controlled. The hydrogen content
of each synthesized powder is determined by using a thermal gravimetric analyzer (TGA). The dehydro-
genation reactions of the synthesized powders are investigated by employing ex situ X-ray diffraction
(XRD), in situ synchrotron XRD and differential thermal analysis (DTA). The results showed that the
incompletely synthesized powder consisted of residual NaAlH4 in the synthesized Mg(AlH4)2, which
demonstrated an initial dehydrogenation temperature as low as 100 ◦C and accompanied with a max-
imum amount (3.1 wt%) of H2 released below 350 ◦C. The mutual catalytic effect of both NaAlH4 and
Mg(AlH4)2 on lowering their initial dehydrogenation temperature is confirmed.
Received 23 December 2011
Received in revised form 7 February 2012
Accepted 9 February 2012
Available online 19 February 2012
Keywords:
Magnesium aluminum hydride
Mechano-chemical activation synthesis
Dehydrogenation reaction
In situ synchrotron X-ray diffraction
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
bond weaker. As a result, the unstable hydrides release hydrogen at
lower temperature. Based on this correlation between ꢀP of M and
Recently, complex metal hydride systems have become a group
of promising hydrogen carriers because of their satisfactory hydro-
gen density and medium hydrogen release temperature [1–4].
Typical complex metal hydrides considered for the hydrogen stor-
age consist of amides, imides, metal borohydrides, and metal
aluminum hydrides (commonly known as metal alanate), etc. For
the commercial metal aluminum hydrides such as LiAlH4 and
NaAlH4, their theoretical gravimetric hydrogen density and first
step dehydrogenation temperature are 10.5 wt%/127–165 ◦C and
7.4 wt%/229–247 ◦C, respectively [1,5]. Obviously, their dehydro-
genation temperature is not low enough for practical PEM-fuel cell
applications on vehicles, even with the utilization of heat from the
operating fuel cell (70–110 ◦C) to facilitate the hydrogen liberation
[6].
the dehydrogenation temperature, magnesium aluminum hydride,
Mg(AlH4)2, is considered to be one of the potential materials for
hydrogen storage.
Since its first synthesis in 1950 by Wiberg and Bauer [10], sev-
eral methods had been developed for the fabrication of Mg(AlH4)2.
Metathesis of NaAlH4 and MgCl2 in diethyl ether (Et2O) or tetrahy-
drofuran (THF) solvent according to reaction (1)
2NaAlH4 + MgCl2 + Et2O/THF → Mg(AlH4)2·Et2O/THF
+ 2NaCl(inEt2O/THF)
(1)
followed by extraction and purification according to reaction (2)
Mg(AlH4)2·Et2O/THF → Mg(AlH4)2 + Et2O/THF
(2)
To lower the dehydrogenation temperature, modification of
complex metal hydride by metal cation substitution is a possible
approach. Nakamori et al. [7–9] indicated that charge transfer from
Mn+ (partial group IA, IIA, and transition metal) to [BH4]− is a sub-
stantial feature for the stability of M(BH4)n, which can be estimated
by Pauling electronegativity (ꢀP) of M. The degree of charge trans-
fer becomes smaller with increasing value of ꢀP, which makes ionic
were reported by Fichtner et al. [11–14]. Mechano-chemical acti-
vation synthesis (MCAS) by a dry ball milling process via reaction
(3), namely
2NaAlH4 + MgCl2 → Mg(AlH4)2 + 2NaCl
(3)
was also reported recently [15–19].
The thermal dehydrogenation behavior of the Mg(AlH4)2 had
also been explored. Claudy et al. [20] proposed that the first
step dehydrogenation of the solvate-free Mg(AlH4)2 initiates at
130 ◦C, while the second one happens at about 300 ◦C. Mamatha
et al. [17,18] indicated the dehydrogenation reactions of Mg(AlH4)2
proceed sequentially in accordance with the following reactions
∗
Corresponding author at: No.1, University Road, Tainan City 701, Taiwan.
Tel.: +886 62757575x62927; fax: +886 62754395.
E-mail address: wttsai@mail.ncku.edu.tw (W.-T. Tsai).
0925-8388/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2012.02.066

C.-H. Yang et al. / Journal of Alloys and Compounds 525 (2012) 126–132
127
(4) and (5), which had been confirmed by different researchers
[12,16,19].
NaAlH4
Mg(AlH4)2
Al
Mg(AlH4)2 → MgH2 + 2Al + 3H2
MgH2 + 2Al → 0.5Mg2Al3 + 0.5Al + H2
(4)
(5)
NaCl
10 hr
Later, Varin et al. [16] found that prolonged milling caused par-
tial dehydrogenation of the synthesized Mg(AlH4)2 based on their
XRD analyses. However, the study was scarce concerning the effect
of synthesis condition on hydrogen desorption behavior, especially
for short milling time. In this study, MCAS was applied to prepare
Mg(AlH4)2 via metal cation substitution from NaAlH4. By alter-
ing the milling time in a relatively wide range as compared with
the study done by Varin et al. [16], the effect of synthesis energy
on the constitutions of the synthesized powders was explored,
and their corresponding dehydrogenation properties and behav-
iors were investigated as well. The optimum hydrogen desorption
performance, such as the highest hydrogen release amount and the
lowest dehydrogenation temperature, was presented.
5 hr
2 hr
1 hr
0.5 hr
0.1 hr
20
30
40
50
60
70
80
Diffraction angle (2θ)
2. Experimental
Fig. 1. XRD patterns of the as-synthesized powders milled for 0.1, 0.5, 1, 2, 5 and
10 h.
2.1. Mechano-chemical activation synthesis of magnesium aluminum hydride
Magnesium aluminum hydride (Mg(AlH4)2) was synthesized using sodium alu-
minum hydride (NaAlH4, Sigma–Aldrich, 90% purity) and anhydrous magnesium
chloride (MgCl2, Sigma–Aldrich, ≥98% purity) as precursors. The precursors were
preserved in a N2-purified glove box, where both the moisture and the oxygen con-
centrations were maintained below 1 ppm. In each metathesis batch, 1 g of mixed
NaAlH4 and MgCl2 powders with a molar ratio of 2:1 was loaded in a 55-ml cylin-
drical vessel made of stainless steel. Specific stainless steel balls were also loaded
into this vessel before sealing tightly. The ball-to-powder weight ratio was 35:1.
Mechano-chemical activation synthesis of Mg(AlH4)2 was performed using a high
energy ball-milling machine (SPEX 8000) for various time, specifically 0.1, 0.5, 1, 2,
5 and 10 h. For the milling time longer than 1 h, milling was executed successively
for 30 min followed by a rest for 15 min.
3. Results and discussion
3.1. Materials characteristics of the synthesized powders
The XRD patterns of the as-synthesized powders prepared with
different milling time are shown in Fig. 1. As depicted in this
figure, the peaks of Mg(AlH4)2 and NaCl were found in all synthe-
sized powders, indicating MCAS via reaction (3) mentioned above
was effective for Mg(AlH4)2 fabrication. For the powder prepared
with a milling time of 0.1 h, the peaks corresponding to NaAlH4
appeared in the XRD pattern, implying the incomplete metathe-
sis reaction. Prolonging milling time to 0.5–2 h, the absence of
the diffraction peaks of the reactants indicated metathesis reac-
tion was completed. However, it was noted that the diffraction
peaks of Al appeared in the XRD pattern and even increased in
intensity with the milling time over 5 h. Moreover, the absence of
the Mg(AlH4)2 peaks with a 10-h milling indicated that complete
dehydrogenation occurred during the prolonged milling process.
Based on the XRD results, the synthesized powders could be clas-
sified into three categories with different constituents, namely,
(1) the hypo-synthesized powder (short milling time) composed
of residual NaAlH4, synthesized Mg(AlH4)2 and NaCl, (2) the
complete-synthesized powder (proper milling time) composed of
Mg(AlH4)2 and NaCl, and (3) the hyper-synthesized powder (over
milling time) contained various amount of Al and NaCl.
2.2. Materials characterization of the synthesized magnesium aluminum hydride
An X-ray diffractometer (XRD, Rigaku MiniFlex II, Cu K␣ radiation) was employed
to identify the crystal structure of the various synthesized powders, before and
after dehydrogenation. In situ powder X-ray diffraction (in situ XRD) was also per-
formed with the aid of Synchrotron Radiation Facility (beamline 01C2 in National
Synchrotron Radiation Research Center in Hsinchu, Taiwan). In each analysis, the
synthesized powder was loaded in a 1-mm diameter glass capillary tube, and then
mounted to the specimen holder. One end of the tube was introduced with a dynamic
N2 gas flow and the other end was open to the atmosphere. During diffraction analy-
sis, the sample was uniformly heated from room temperature to 365 ◦C at a ramping
rate of 5 ◦C min−1 by blowing hot air outside the capillary tube. The wavelength of
˚
the synchrotron X-ray was 1.033209 A. Every 2-D diffraction pattern was succes-
sively collected by a Mar345 imaging plate. Then, the 2-D diffraction pattern was
converted to 1-D pattern by the Fit2D software. Accordingly, the high temperature
transition of crystal structure of the synthesized powders was realized. Besides, the
morphology of the synthesized powders was examined using a scanning electron
microscope (SEM, Hitachi SU-1500). Elements distribution was characterized using
the energy dispersive spectrometer (EDS) equipped on SEM.
Over-milling caused the generation of excessive heat from the
fast impact of the steel balls on the powder precursors and the
vessel wall, resulting in the increase of temperature above the
decomposition temperature of the synthesized Mg(AlH4)2. Conse-
quently, the premature dehydrogenation of Mg(AlH4)2 occurred.
Similar observation has been reported by Varin et al. [16], who
focused on the synthesis conditions in the range of complete-
to hyper-synthesized Mg(AlH4)2 and the corresponding heat flow
events during dehydrogenation reaction. In this present study, the
dehydrogenation behavior of the hypo-synthesized powder pre-
pared with a shorter milling time was also explored. As will be
discussed later, the residual precursors would affect the dehydro-
genation performance.
2.3. Thermal decomposition and dehydrogenation performance
Differential thermal analyses (DTA) of the milled powders were performed using
a NETZSCH STA 409 PC analyzer. In each test, 80 mg of the synthesized powder was
loaded in an alumina crucible. The measurement was conducted in an argon gas
flow at a rate of 70 ml min−1, and sample heating from room temperature to 350 ◦
C
at a rate of 5 ◦C min−1
.
Thermogravimetric analysis (TGA) using a high-pressure microbalance (Cahn
D-110) was conducted to evaluate the dehydrogenation behavior of the synthe-
sized powders. The amount of H2 release and the dehydrogenation temperature
were of particular interest. In each test, the synthesized powder with an initial
weight of ca. 500 mg was loaded in a quartz crucible and transferred into the high-
pressure microbalance chamber. Then, the chamber was evacuated to 1 × 10−4 torr
followed by the introduction of H2 gas (99.999% purity) to ambient pressure. Until
the microbalance system was stable, the TGA test from room temperature to 350 ◦
at a heating rate of 4–5 ◦C min−1 was executed and recorded.
C
SEM micrographs showing the morphologies of the raw reac-
tant powders, specifically NaAlH4 and MgCl2, are demonstrated in

128
C.-H. Yang et al. / Journal of Alloys and Compounds 525 (2012) 126–132
Fig. 2. SEM micrographs of raw reactant powders: (a) NaAlH4, (b) MgCl2 and the various as-synthesized powders milled for (c) 0.5 h, (d) 10 h.
Fig. 2(a) and (b). The particle size of NaAlH4 powder was relatively
uniform, while that of flake-shaped MgCl2 powder varied in a wide
range from sub-micrometer to hundreds of micrometer. Fig. 2(c)
and (d) shows the SEM micrographs of the powders synthesized for
0.5 and 10 h, respectively. The particle size of the synthesized pow-
ders was significantly reduced by comparing the SEM micrographs
shown in Fig. 2(c) and (d) with those shown in Fig. 2(a) and (b).
Extended milling from 0.5 to 10 h, however, did not cause further
reduction in particle size as demonstrated in Fig. 2(d). Fig. 3 shows
the elements distribution of 0.5 h-synthesized powder, examined
by EDS analysis. The distribution of the main elements including
Mg, Na, Al and Cl was uniform according to the EDS mapping results
shown in Fig. 3. The detection of O indicated that oxidation on the
surface of the active Mg(AlH4)2 was inevitable during MCAS.
synthesized powders prepared by MCAS with various milling time,
the variation of weight change with temperature (from room tem-
perature to 350 ◦C) for each powder is demonstrated in Fig. 5. As
shown in this figure, the shorter the milling time was, the more
H2 was released from the synthesized powder. The temperature
at which the synthesized powders started to dehydrogenate also
depended on the period of milling time. As shown in Fig. 5, the dehy-
drogenation commenced at about 100 ◦C for the 0.1 h-synthesized
powder, while that for the 0.5, 1, 2 h-synthesized powders was
about 170 ◦C. The dehydrogenation reaction was almost not visible
for the 10 h-synthesized powder. The result for as-received NaAlH4
was also included in this figure for comparison. It was found that
NaAlH4 started to dehydrogenate at about 210 ◦C, which was higher
than that of the synthesized Mg(AlH4)2.
It was noticed that the precursors did not completely reacted
under the short milling time such as 0.1 h. As a result, the
hypo-synthesized powder still contained some residual NaAlH4 as
revealed in the XRD pattern shown in Fig. 1. The lower dehydro-
genation temperature observed for the 0.1 h-synthesized powder,
as revealed in Fig. 5, suggested that NaAlH4 could destabilize
Mg(AlH4)2 and activate its decomposition at much lower temper-
ature. Srivastava et al. [21] had explored the synergistic effect of
Mg(AlH4)2 on accelerating the dehydrogenation kinetics and low-
ering the dehydrogenation temperature of NaAlH4. In our case,
however, the residual NaAlH4 acted as a promoter for the decom-
position of Mg(AlH4)2. Further investigation on the synergistic
behavior of dehydrogenation reaction of Mg(AlH4)2 and NaAlH4
mixed hydride system was performed with the assistance of in situ
synchrotron XRD analysis, as will be discussed later.
3.2. Thermal decomposition and dehydrogenation behavior
Thermal decomposition behavior of various synthesized pow-
ders was explored by conducting DTA analysis. The results
demonstrated in Fig. 4 were obtained by heating the powders
from room temperature to 350 ◦C at a ramping rate of 5 ◦C min−1
in an argon gas flow of 70 ml min−1. For the MCAS powders pre-
pared with a milling time less than 2 h, two endothermic peaks
in the temperature range of 160–180 ◦C and 250–265 ◦C, respec-
tively, appeared in each DTA curve. These two endothermic peaks
corresponded to the two dehydrogenation reactions. For the 5,
10 h-synthesized powders, however, no peaks were seen in the
DTA curves. In these two powders, prolonged milling caused
over-synthesis, resulting in the premature dehydrogenation as evi-
denced in the XRD patterns shown in Fig. 1.
The total amounts of H2 released, based on TGA analysis (Fig. 5),
were 3.1 wt%, 2.75 wt%, 2.45 wt%, 2.45 wt% and 0.92 wt% from the
powders synthesized by milling for 0.1, 0.5, 1, 2 and 5 h, respec-
tively. The weight change results definitely indicated that the
Quantitative evaluation in determining the amount of H2 des-
orption and the dehydrogenation temperature of the synthesized
powders was performed by thermogravimetric analysis. For the

C.-H. Yang et al. / Journal of Alloys and Compounds 525 (2012) 126–132
129
Fig. 3. EDS mappings showing the distributions of Mg, Na, Al, Cl and O in 0.5 h-synthesized powder.
1.2
0.8
0.4
0
milling time (h)
10
5
100
98
2
1
0.5
milling time (h)
10
5
2
1
0.5
0.1
96
As-received NaAlH4
0
100
200
300
0
100
200
300
Temperature (oC)
Temperature (oC)
Fig. 5. Thermogravimetric analyses of NaAlH4 and the as-synthesized powders
milled for 0.1, 0.5, 1, 2, 5 and 10 h heated from room temperature to 350 ◦C (ramping
rate: 5 ◦C min−1) under ambient H2 gas.
Fig. 4. DTA of the as-synthesized powders milled for 0.5, 1, 2, 5 and 10 h heated
from room temperature to 350 ◦C (ramping rate: 5 ◦C min−1) in an argon gas flow of
70 ml min−1
.

130
C.-H. Yang et al. / Journal of Alloys and Compounds 525 (2012) 126–132
Al
Mg2Al3
NaCl
5 hr
2 hr
1 hr
0.5 hr
0.1 hr
20
30
40
50
60
70
80
Diffraction angle (2θ)
Fig. 6. XRD patterns of the as-synthesized powders milled for 0.1, 0.5, 1, 2 and 5 h
after thermal dehydrogenation at 350 ◦C.
powder prepared with shorter milling time could release higher
amount of hydrogen. Besides, the amounts of H2 released from the
second step dehydrogenation reaction were very small as com-
pared with that of the first step reaction. As mentioned above,
incomplete MCAS process would leave the reactants, such as
NaAlH4, in the synthesized powder. Over-synthesis with prolonged
milling time would lead to premature H2 release of the Mg(AlH4)2
complex metal hydride. Clearly, the difference in the amounts of
H2 released from the synthesized powders, as measured in TGA
analysis, was mainly attributed to the final composition of pow-
ders after MCAS process. It was also noted that these values were
lower than the theoretical capacity of the fully synthesized powders
containing Mg(AlH4)2 and NaCl (3.9 wt%). Even for the complete-
synthesized Mg(AlH4)2 powders milled for 0.5–2 h, the amounts
of H2 release were about 62–70 wt% of the theoretical H2 density.
Beside the incomplete dehydrogenation, oxidation of Mg(AlH4)2 in
the course of MCAS process might be also responsible for the loss
of H2 generation.
The XRD patterns of various synthesized powders after TGA test
(up to 350 ◦C) are shown in Fig. 6. In each pattern, NaCl was iden-
tified, which was the main product of MCAS process, other than
Mg(AlH4)2. The diffraction pattern of the 0.1 h-synthesized pow-
der revealed the presences of metallic Al and Mg2Al3, which were
resulted from the thermal dehydrogenation of Mg(AlH4)2. In the
referred studies on the dehydrogenation of NaAlH4 powder by sev-
eral other researchers [22–24], Al and NaH had been identified after
TGA test. The authors of these investigations concluded that NaAlH4
would decompose to Na3AlH6, Al, NaH and gaseous H2 during heat-
ing to 350 ◦C. However, none of these species were identified in
the XRD patterns shown in Fig. 6. For the 0.5 and 1 h-synthesized
powders, the diffraction patterns showed that the dehydrogenated
powders consisted of Al and intermetallic Mg2Al3 alloy. The pres-
ence of Mg2Al3 was also originated from the dehydrogenation
reaction. In the diffraction patterns of the 2 and 5 h-synthesized
powders, only Al peaks, beside those of NaCl, were observed after
TGA test. The absence of the peaks of Mg(AlH4)2 in each XRD pattern
indicated that complete dehydrogenation occurred after heating
the powders to 350 ◦C.
Fig. 7. (a) In situ synchrotron XRD patterns of the as-received NaAlH4 heated from
room temperature to 365 ◦C; and (b) variation of the strongest peak intensity with
temperature for the major species appeared during thermal treatment.
from room temperature up to 365 ◦C to induce the dehydrogenation
reactions. The diffraction patterns were then collected every 10 or
20 ◦C and compiled to give a result shown in Figs. 7(a), 8(a) and 9(a).
For the as-received NaAlH4, as shown in Fig. 7(a), the presence
of Na3AlH6 peaks accompanied with the decreasing intensity
of NaAlH4 peaks was observed at temperature higher than
175 ◦C. This change was associated with the first step dehy-
drogenation reaction, namely NaAlH4 → 1/3Na3AlH6 + 2/3Al + H2.
Continued heating to about 275 ◦C, the formation of NaH as
well as the gradually decreasing intensity of Na3AlH6 indi-
cated the onset of the second step dehydrogenation reaction
as 1/3Na3AlH6 → NaH + 1/3Al + 1/2H2. At temperature higher than
305 ◦C, the dehydrogenated species, namely NaH and Al, initially
oxidized to form NaAlO2. The progressive transformation of reac-
tion products in view of their relative intensities obtained from
in situ XRD analyses are shown in Fig. 7(b). The effect of tempera-
ture on dehydrogenation reaction is clearly demonstrated.
The in situ synchrotron XRD patterns of the MCAS powder con-
taining Mg(AlH4)2, prepared by milling for 0.5 h and heating from
room temperature to 365 ◦C are shown in Fig. 8(a). At tempera-
ture below 145 ◦C, Mg(AlH4)2 was stable and no significant change
was observed. Increasing temperature to 165 ◦C, the presence of
MgH2 and Al peaks and the decreasing intensity of Mg(AlH4)2 indi-
cated the occurrence of the first step dehydrogenation reaction,
described as Mg(AlH4)2 → MgH2 + 2Al + 3H2, basically in agreement
with reaction (4) reported by others [12,16–19]. The dependence
of the relative intensity on temperature for the major phases
Thermal decomposition behavior of as-received NaAlH4 and
MCAS-prepared Mg(AlH4)2 was studied by in situ synchrotron
X-ray diffraction, respectively. During in situ synchrotron X-ray
diffraction analysis, the powders were heated at a rate of 5 ◦C min−1

C.-H. Yang et al. / Journal of Alloys and Compounds 525 (2012) 126–132
131
Fig. 8. (a) In situ synchrotron XRD patterns of the 0.5 h-synthesized powder heated
from room temperature to 365 ◦C; and (b) variation of the strongest peak intensity
with temperature for the major species appeared during thermal treatment.
Fig. 9. (a) In situ synchrotron XRD patterns of the 0.1 h-synthesized powder heated
from room temperature to 365 ◦C; and (b) variation of the strongest peak intensity
with temperature for the major species appeared during thermal treatment.
is also depicted in Fig. 8(b). At 205 ◦C, the peaks corresponding
to Mg(AlH4)2 disappeared, indicating the completion of the first
step dehydrogenation. The steady peak intensities of MgH2 and Al
also indicated the occurrence of the first step dehydrogenation.
Continued heating to the temperature range of 225–305 ◦C, the
peak intensity of MgH2 gradually reduced, indicating the occur-
rence of the second step dehydrogenation, mostly described as
MgH2 + 2Al → 0.5Mg2Al3 + 0.5Al + H2. The absence of Mg2Al3 peaks
in Fig. 8(a) was due to the small amount of loading used in in situ
XRD analysis. However, according to the XRD patterns of 0.5 h-
synthesized powder after TGA measurement at 350 ◦C as revealed
in Fig. 6, the reaction of MgH2 and Al to form Mg2Al3 was con-
firmed. Throughout the whole temperature scanning, NaCl as the
by-product of MCAS was identified and remained stable without
change.
The in situ synchrotron XRD patterns of the MCAS powder con-
taining Mg(AlH4)2 and NaAlH4, prepared by milling for 0.1 h and
heating from room temperature to 365 ◦C are shown in Fig. 9(a).
Since the 0.1 h-synthesized powder was a mixture consisting of
Mg(AlH4)2, residual NaAlH4 and NaCl, the lower temperature XRD
patterns revealed their contribution as shown in Fig. 9(a). Increas-
ing the temperature to 105 ◦C, no significant change with regard to
phase transformation was observed. As the temperature was raised
to 125 ◦C, the peaks associated with Al and MgH2 began to appear,
indicating the initiation of the first step dehydrogenation reaction
like that of 0.5 h-synthesized powder. It was likely that the initial
dehydrogenation temperature was between 105 and 125 ◦C when
the synthesized powder was heated up. By comparing with that of
the completely synthesized powder (with 0.5 h milling time), the
initial dehydrogenation temperature was about 60 ◦C lower with
the co-existence of Mg(AlH4)2 and NaAlH4. The latter seemed to
behave as a catalyst for the dehydrogenation of Mg(AlH4)2.
The variation of the relative intensity with temperature for the
major species is demonstrated in Fig. 9(b). The detection of MgH2
and Al peaks at temperature as low as 105 ◦C indicated that decom-
position of Mg(AlH4)2 was assisted by the presence of NaAlH4. The
absence of Mg(AlH4)2 peaks at about 165 ◦C again suggested the
completion of the first step dehydrogenation of Mg(AlH4)2. The fact
that the intensity of NaAlH4 peaks started to decrease at temper-
ature below 105 ◦C, as shown in Fig. 9(b), might also suggest that
Mg(AlH4)2 could assist dehydrogenation of NaAlH4.
Continued heating the 0.1 h-synthesized powder to about
265 ◦C, the diffraction patterns revealed that Al and MgH2, beside
NaCl, were the main phases present. As the temperature was raised
above 285 ◦C, MgH2 and Al gradually decreased in their intensi-
ties, indicating the occurrence of the second step dehydrogenation
via MgH2 + 2Al → 0.5Mg2Al3 + 0.5Al + H2 reaction. The presence of
Mg2Al3 peaks in the diffraction patterns above 285 ◦C confirmed
this reaction. The dehydrogenation behavior of mixed NaAlH4 and
Mg(AlH4)2 powders was different from their single constituent. The
synergistic and mutual catalytic effects of them, not only on lower-
ing the initial dehydrogenation temperature of mixed hydrides but
also on releasing more amount of H2 than the individual Mg(AlH4)2,
were noticed. The mutual catalytic effect on both NaAlH4 and
Mg(AlH4)2 in dehydrogenation reaction needs further investiga-
tion.

132
C.-H. Yang et al. / Journal of Alloys and Compounds 525 (2012) 126–132
4. Conclusions
[4] S. Satyapal, J. Petrovic, C. Read, G. Thomas, G. Ordaz, Catal. Today 120 (2007)
246–256.
[5] W. Grochala, P.P. Edwards, Chem. Rev. 104 (2004) 1283–1315.
[6] J. Larminie, A. Dicks, Fuel Cell Systems Explained, Second ed., John Wiley and
sons, Chichester, UK, 2003.
[7] Y. Nakamori, K. Miwa, A. Ninomiya, H.W. Li, N. Ohba, S. Towata, A. Zuttel, S.
Orimo, Phys. Rev. B 74 (2006) 045126.
[8] Y. Nakamori, H.W. Li, K. Kikuchi, M. Aoki, K. Miwa, S. Towata, S. Orimo, J. Alloys
Compd. 446–447 (446) (2007) 296–300.
[9] Y. Nakamori, H.W. Li, M. Matsuo, K. Miwa, S. Towata, S. Orimo, J. Phys. Chem.
Solids 69 (2008) 2292–2296.
[10] E. Wiberg, R. Bauer, Z. Naturforsch. 5 (1950) 396–397.
[11] M. Fichtner, O. Fuhr, J. Alloys Compd. 345 (2002) 286–296.
[12] M. Fichtner, O. Fuhr, O. Kircher, J. Alloys Compd. 356–357 (356) (2003) 418–420.
[13] M. Fichtner, J. Engel, O. Fuhr, A. Gloss, O. Rubner, R. Ahlrichs, Inorg. Chem. 42
(2003) 7060–7066.
[14] K. Komiya, N. Morisaku, Y. Shinzato, K. Ikeda, S. Orimo, Y. Ohki, K. Tatsumi, H.
Yukawa, M. Morinaga, J. Alloys Compd. 446–447 (446) (2007) 237–240.
[15] R.A. Varin, Ch. Chiu, T. Czujko, Z. Wronski, Nanotechnology 16 (2005)
2261–2274.
Mg(AlH4)2 was prepared by a MCAS process employing NaAlH4
and MgCl2 as precursors. The synthesis was completed with a
milling time controlled at 0.5–2 h. The results of HPTGA analyses
showed that the maximum amount about 3.1 wt% of H2 could be
released from the MCAS powder which consisted of Mg(AlH4)2 and
residual NaAlH4. Over milling for a time greater than 2 h caused
premature dehydrogenation from the MCAS powders. In situ
synchrotron XRD analysis confirmed the two-dehydrogenation
reaction of Mg(AlH4)2. Both HPTGA and in situ XRD analyses
showed that the presence of NaAlH4 could assist dehydrogenation
of Mg(AlH4)2 by lowering the initial dehydrogenation temperature
close to around 100–105 ◦C, which was about 60–70 ◦C lower than
that of plain Mg(AlH4)2 hydride.
[16] R.A. Varin, Ch. Chiu, T. Czujko, Z. Wronski, J. Alloys Compd. 439 (2007) 302–311.
[17] M. Mamatha, B. Bogdanovic, M. Felderhoff, A. Pommerin, W. Schmidt, F. Schuth,
C. Weidenthaler, J. Alloys Compd. 407 (2006) 78–86.
Acknowledgement
[18] M. Mamatha, C. Weidenthaler, A. Pommerin, M. Felderhoff, F. Schuth, J. Alloys
Compd. 416 (2006) 303–314.
[19] Y. Kim, E.K. Lee, J.H. Shim, Y.W. Cho, K.B. Yoon, J. Alloys Compd. 422 (2006)
283–287.
[20] P. Claudy, B. Bonnetot, J.M. Letoffe, J. Therm. Anal. Calorim. 15 (1979) 119–128.
[21] M. Sterlin Leo Hudson, D. Pukazhelvan, G. Irene Sheeja, O.N. Srivastava, Int. J.
Hydrogen Energy 32 (2007) 4933–4938.
[22] P. Claudy, B. Bonnetot, G. Chahine, J.M. Letoffe, Thermochim. Acta 38 (1980)
75–88.
The authors would like to thank the National Science Council of
Republic of China (Taiwan) for financial supports under grant NSC
97-2221-E-006-113-MY3.
References
[1] D. Chandra, J.J. Reilly, R. Chellappa, JOM 58 (2006) 26–32.
[2] S. Orimo, Y. Nakamori, J.R. Eliseo, A. Zuttel, C.M. Jensen, Chem. Rev. 107 (2007)
4111–4132.
[3] B. Sakintuna, F. Lamari-Darkrim, M. Hirscher, Int. J. Hydrogen Energy 32 (2007)
1121–1140.
[23] K.J. Gross, S. Guthrie, S. Takara, G. Thomas, J. Alloys Compd. 297 (2000)
270–281.
[24] L. Zaluski, A. Zaluska, J.O. Strom-Olsen, J. Alloys Compd. 290 (1999) 71–78.