Hydrogen reactivity of Li-containing hydrogen storage materials

Article abstract of DOI:10.1016/S0925-8388(03)00357-8

In hydrogen energy systems, the hydrogen storage alloy is expected as one of important materials for energy storage and transportation. However, they weigh so much for sufficient amounts of hydrogen, therefore, light hydrogen storage materials with much higher storage capacities are required. Alkaline complex hydrides are promising candidates as light hydrogen storage materials. In this study we tried to prepare a sample based on the Na-Li-Al system using the dry process of ball-milling as mechanical synthesis and investigate their hydrogen reaction characteristics. The same study for the sample with 10 wt.% La₂O₃ was also carried out and its hydrogen reactivity compared. The starting materials were LiAlH₄, NaH. They were ball-milled with hexane dehydrated in 1 bar Ar. The hydrogen reactivity of the samples obtained were measured by the high-pressure Sievert's type apparatus. For the sample of NaH+LiH+Al, about 1.9～2.0 wt.% of hydrogen was absorbed reversibly within 150 min. The addition of 10 wt.% La₂O₃ improved both the ab/desorption reaction rate and cyclable hydrogen capacity.

Full text of DOI:10.1016/S0925-8388(03)00357-8

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Journal of Alloys and Compounds 356–357 (2003) 358–362
L
www.elsevier.com/locate/jallcom
Hydrogen reactivity of Li-containing hydrogen storage materials
a
a
R. Genmaa, H.H. Uchidab , , N. Okada , Y. Nishi
*
aDepartment of Materials Science, Tokai University, Kitakaname 1117, Hiratsuka, Kanagawa 259-1292, Japan
bDepartment of Human Development, Tokai University, Kitakaname 1117, Hiratsuka, Kanagawa 259-1292, Japan
Received 1 September 2002; received in revised form 15 December 2002; accepted 6 February 2003
Abstract
In hydrogen energy systems, the hydrogen storage alloy is expected as one of important materials for energy storage and transportation.
However, they weigh so much for sufficient amounts of hydrogen, therefore, light hydrogen storage materials with much higher storage
capacities are required. Alkaline complex hydrides are promising candidates as light hydrogen storage materials. In this study we tried to
prepare a sample based on the Na–Li–Al system using the dry process of ball-milling as mechanical synthesis and investigate their
hydrogen reaction characteristics. The same study for the sample with 10 wt.% La2O3 was also carried out and its hydrogen reactivity
compared. The starting materials were LiAlH4, NaH. They were ball-milled with hexane dehydrated in 1 bar Ar. The hydrogen reactivity
of the samples obtained were measured by the high-pressure Sievert’s type apparatus. For the sample of NaH1LiH1Al, about 1.9|2.0
wt.% of hydrogen was absorbed reversibly within 150 min. The addition of 10 wt.% La2O3 improved both the ab/desorption reaction rate
and cyclable hydrogen capacity.
2003 Elsevier B.V. All rights reserved.
Keywords: Hydrogen storage; Alanates; Mechanical alloying; Lanthanum oxide
1. Introduction
been also explored. The preparation process based on
reactions with the use of solvents needs, however, some
complicated steps of purification and filtration. Additional
severe conditions of high pressure of hydrogen above 100
atm and higher temperature are needed for the activation
(1 atm5101325 Pa). A more simplified preparation pro-
cess is, therefore, desired. From this point of view, some
studies on fabricating alkali metal complex hydrides using
mechano-chemical synthesis have been reported [5,6].
In this study, we have taken Li as a base metal, which is
the lightest metal to form a metal hydride, and tried to
prepare the sample based on the non-stoichiometric Na–
Li–Al system without any catalytic dopant by ball milling.
Some La2O3 powder was also given during the milling.
Their hydrogen reactivity was investigated using the high-
pressure Sieverts’ type apparatus.
In hydrogen energy system, hydrogen storage alloys are
expected to play some important roles, such as hydrogen
storage and transportation. However, they weigh so much
for sufficient amounts of hydrogen absorbed, therefore,
lightweight hydrogen storage materials with higher storage
capacities are required.
´
In 1997, Bogdanovic and Schwickardi [1] reported some
new hydrogen storage materials of Ti-doped alkali metal
complex hydrides, NaAlH4, Na3AlH6 and Na2LiAlH6
prepared by liquid reactions. Especially Ti-doped
Na2LiAlH6, absorbed more than about 2.6 wt.% of hydro-
gen at 484 K, under 4.0 MPa H2 successfully. The plateau
pressure of Ti doped Na2LiAlH6 is about 1.4 MPa, which
is about 2.0 MPa lower than that of Na3AlH6. After this
report, many studies on NaAlH4 and Na3AlH6 as hydrogen
storage materials have been carried out from various points
of view such as reaction kinetics [2], crystallographic
analysis [3], microstructure examination [4] and so on.
Sample preparation and catalyst-doping processes have
2. Experimental
The starting materials were NaH and LiAlH4 (Wako
Pure Chemical Co. Ltd.), and La2O3 (High Purity Chemi-
cal Co. Ltd.). At first, NaH was washed by filtration with
hexane dehydrated to clear off the preservation mineral oil.
Sample materials were weighed for the composition of
*
Corresponding author. Tel.: 181-463-581-211x3413; fax: 181-463-
502-208.
E-mail address: hhuchida@keyaki.cc.u-tokai.ac.jp (H.H. Uchida).
0925-8388/03/$ – see front matter
doi:10.1016/S0925-8388(03)00357-8
2003 Elsevier B.V. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
R. Genma et al. / Journal of Alloys and Compounds 356–357 (2003) 358–362
359
Table 1
Ball milling conditions
Mill type
Material of vial/balls
Capacity of vial
Ball diameter
and the number
Total weight of the sample to be milled
Capacity ratio of vial vs. balls
Weight ratio of balls vs. sample
Dispersion medium
Rotary mill
SUS 316L/SUS 304
70.33 cm3
3530.95 cm
2530.60 cm
1.5 g
1.7:1
70:1
Hexane dehydrated
(,0.3 ppm H2O)
Ar (99.999%)
180 rpm
Milling atmosphere
Rotating velocity
Milling period
30 h
NaH:LiAlH451:1 (NaLiAlH5) and additionally 10 wt.%
of La2O3, then ball-milled under the same conditions as
shown in Table 1. The mill we used was laboratory-scale
one, which was a rotary mill from Irie Shokai Co. Ltd., not
a planetary one. The milled samples obtained of 0.2 g each
were loaded in the reaction bed attached with a stop valve
under a glove box. All the processes mentioned above
were carried under an inert gas atmosphere with low
moisture content in a glove box. Their hydrogen reactivity
was measured after setting on a high-pressure Sieverts’
type apparatus. To confirm the phases and structure of the
sample, X-ray diffraction (XRD) measurements were
carried out by Miniflex (Rigaku Co.) with a Cu Ka X-ray
source. Microstructural observation was carried out using
scanning electron microscopy (SEM; Hitachi S-3200N1
Horiba EMAX5771W).
Fig. 1. The first dehydrogenation curve of the sample without La2O3.
gen amount in the sample induced by collision during ball
milling, the ratio of released hydrogen amount does not
correspond to the reaction expressed above, that is, some
of hydrogen was possibly released during milling, while
the ratio of hydrogen amount derived from the first step
Because each sample obtained was hydride, before the
absorption and desorption measurements, they were initial-
ly dehydrogenated by heating up to 543 K under a 0.15
MPa He atmosphere and then evacuated to 0.1 Pa. After
the dehydrogenation, samples were cooled to 484 K under
0.3 MPa of He.
3. Results and discussion
The first dehydrogenation curve of the sample without
La2O3 and XRD results are shown in Figs. 1 and 2,
respectively. During heating up from room temperature to
543 K, the curve showed two steps. The first step began
from 423 K, and the second from 483 K, which correspond
to well-known thermal decomposition as shown as follows:
NaAlH4↔1/3Na3AlH6 1 2/3Al 1 H2
1/3Na3AlH6↔NaH 1 1/3Al 1 1/2H2
(1)
(2)
In spite of the fact that the starting materials were NaH
and LiAlH4, aiming for the composition of NaLiAlH5, this
result suggests that the milled product contains mainly
NaAlH4 phase. However, if considering the loss of hydro-
Fig. 2. X-Ray diffraction patterns for the sample without La2O3; (a) as
ball milled, (b) after the first dehydrogenation at 473 K, (c) after the first
dehydrogenation at 543 K, (d) after hydrogenation at 484 K.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
360
R. Genma et al. / Journal of Alloys and Compounds 356–357 (2003) 358–362
and second step was not 2:1. The XRD results (Fig. 2a)
also suspend this consideration. In the case of the sample
with 10 wt.% La2O3, first dehydrogenation was yielded in
the same manner.
After the first dehydrogenation, hydrogenation measure-
ments were carried out under the same conditions at 484 K
and 3.9 MPa of applied hydrogen pressure. Fig. 3 shows
the first hydrogenation curves for the sample with and
without La2O3. The hydrogen amounts absorbed for the
sample with and without La2O3 were 1.98 wt.% each,
which corresponds to H/NaH1LiH1Al51.13. For the
sample with 10 wt.% La2O3, the absorbed amount is
expressed without the weight of La2O3. Comparing the
curves of each sample, the initial reaction rate of the
sample with 10 wt.% La2O3 is higher than that without
La2O3.
After removing the residual hydrogen from the reaction
bed, 0.15 MPa of He was introduced subsequently, then H2
desorption measurement was carried out. Fig. 4 shows the
second desorption curves for the sample with and without
La2O3 at 484 K under 0.15 MPa of He atmosphere. The
sample with 10 wt.% La2O3 also showed better kinetics
than that without La2O3 in dehydrogenation as it did in
hydrogenation.
The results of XRD measurements for the sample
without La2O3 at each stage of ‘‘As ball milled’’, ‘‘After
the first dehydrogenation at 473 K’’, ‘‘After the first
dehydrogenation at 543 K’’ and ‘‘After the first hydro-
genation’’ are shown in Fig. 2. As previously mentioned,
milled products contained some different hydride phases,
such as NaAlH4 and Na2LiAlH6 (see Fig. 2a). Some of
their decomposition products, NaH, LiH, and Al react
during the first dehydrogenation resulting in Na2LiAlH6
phase (see Fig. 2b and c). As some of the reasons why
there are no obvious peaks containing lithium, e.g. LiH in
Fig. 4. The second dehydrogenation curves of the sample with and
without La2O3.
Fig. 2a, it should be considered that the decrease of grain
size and introduction of defects by ball milling [7]. These
factors could weaken the peak intensity and broaden the
peak width. Therefore it is possible to suggest that LiH
exists behind the background as a result of these effects. It
can be followed by a result that the observed peak width of
NaAlH4 also seems to be relatively wide. The reappear-
ance of LiH phase in Fig. 2b can be due to an effect of
recovery and recrystallization caused by subsequent heat-
ing.
Fig. 5 shows a SEM image of the sample powder ball
milled with 10 wt.% La2O3. The white color indicates the
existence of La2O3. An almost homogeneous dispersion of
La2O3 caused by ball milling was confirmed from the
image.
A remarkable effect by addition of La2O3 was shown in
cycling behavior. Fig. 6 shows the hydrogen capacity in
the course of four de/rehydrogenation open cycles and its
Fig. 3. The first hydrogenation curves of the sample with and without
La2O3.
Fig. 5. SEM image of the sample powder ball milled with 10 wt.%
La2O3, shown as white color.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
R. Genma et al. / Journal of Alloys and Compounds 356–357 (2003) 358–362
361
Fig. 6. Comparison of hydrogen capacity and reaction rate in the course of four re/dehydrogenation open cycles for the sample with and without La2O3.
initial reaction rate at each reaction. After the fourth cycle,
the hydrogen capacity at hydrogenation decreased to 86%
for the sample without La2O3, while the sample with 10
wt.% La2O3 remained at 98% of hydrogen capacity. A
similar effect was also shown in the initial reaction rate.
From the pressure dependence of initial reaction rate, the
rate determining step can be investigated [8]. The depen-
dence coefficient; n of v~pn indicates the rate determining
step of the molecular dissociation (n51) or permeation/
diffusion of atomic hydrogen (n50.5). For samples with
and without La2O3, H2 dissociation seems to be the rate
determining step as shown in Fig. 7. Additionally, the
sample with La2O3 shows a higher dissociation rate, i.e.,
La2O3 seems to act as a catalytic substance in alkali-Al
hydrogenation. The catalytic effect of La2O3 was also
reported elsewhere [9].
4. Conclusion
In this paper the hydrogen reactivity of alkali metal
complex hydrides prepared by ball milling NaH1LiAlH4
as the starting material, and the effect of the addition of
La2O3 powder were investigated. The hydrogen capacity
of the sample at 484 K and 3.9 MPa of hydrogen pressure
was about 1.9|2 wt.%, which decreased to 1.7 wt.% at
several cycles without La2O3 powder. This small value is
caused by the absorption reaction of Na2LiAlH6. This
phase was confirmed by XRD, although initially we
intended a non-stoichiometric Na–Li–Al system for ob-
taining higher concentration of H. The addition of La2O3
resulted in enhancing the dissociative reaction rate and
keeps hydrogen capacity and reaction rate in the sorption
cycles.
References
´
[1] B. Bogdanovic, M. Schwickardi, J. Alloys Comp. 253–254 (1997)
1–9.
[2] G.J. Thomas, S.E. Guthrie, K. Gross, Proceedings of the 1999 US
DOE Hydrogen Program Review, NREL/CP-570-26938.
[3] K.J. Gross, S. Guthrie, S. Takara, G. Thomas, J. Alloys Comp. 297
(2000) 270–281.
Fig. 7. Pressure dependence of initial reaction rate for the sample (a) with
10 wt.% La2O3 and (b) without La2O3.