Materials Transactions, Vol. 50, No. 7 (2009) pp. 1855 to 1858
#2009 The Japan Institute of Metals
Optimum Hydrogen Desorption Properties in LiH-LiOH Composites
Masatsugu Kawakami1; * , Takahiro Kuriiwa1 , Atsunori Kamegawa1 ,
Hitoshi Takamura1 , Masuo Okada1 and Tomohiro Kaburagi2
1
2
Department of Materials Science, Graduate School of Engineering, Tohoku University, Sendai 980-8579, Japan
Nissan Motor Co., Ltd., Yokosuka 237-8523, Japan
The effect of variation of LiOH content on hydrogen desorption properties in LiH-(30–60) mol%LiOH composites was investigated. The
addition of LiOH destabilized LiH in desorbing hydrogen below 300 C for all composites while pure LiH desorbs hydrogen above 650 C. The
hydrogen desorption temperature of these composites decreased with decreasing the content of LiOH. The onset temperature of hydrogen
desorption lowered to 262 C for the sample of LiH-30 mol%LiOH.
In the TDS measurement, the generation of water was observed around 420 C for the samples of LiH-(40–60) mol%LiOH due to
decomposition of unreacted LiOH. The intensity of the water peak from these composites in TDS decreases with decreasing the content of LiOH.
The water generation was unobserved from LiH-30 mol%LiOH composite. These results indicate that LiH-(30 and 40) mol%LiOH is a suitable
composition for hydrogen desorption in this study. [doi:10.2320/matertrans.M2009055]
(Received February 13, 2009; Accepted April 21, 2009; Published June 10, 2009)
Keywords: hydrogen storage materials, solid-solid reactions, LiH-LiOH composites, lowering desorption temperature
1.
Introduction
2LiH þ NaOH ! Li2 O þ NaH þ H2
Lightweight materials such as used for hydrogen storage
media or on-board type hydrogen generator are required
for fuel cells in vehicles or portable electronic devices.
Then, some light metal hydrides such as LiH and MgH2
are paid much attention with its high hydrogen storage
capacity. The hydrogen storage capacity for LiH and MgH2
are theoretically calculated to be 12.7 mass% and
7.7 mass%, respectively. However, LiH melts at 689 C
without hydrogen desorption.1) Its desorption temperature is
higher than its melting point. So, it might be inconvenient
for practical use as hydrogen source. On the other hand,
MgH2 is well known to generate hydrogen about 300 C
under ambient pressure.2,3) Working temperature of both
materials is high and which caused major difficulty for
practical on-board use of these materials. The other major
problems for these materials for practical utilization are
its large reaction of heat for hydrogen desorption which
requires hydrogen storage system for large capacity of
heater.
Then, the destabilization of these ionic hydrides for
decrement of working temperature as well as reduction of
heat for hydrogen generation has been attracting strong
interests as for use for on-board type fuel cells and so on.
J. J. Vajo et al. reported the destabilization of LiH or
MgH2 by additives such as pure Si,4) some hydroxides5) and
LiBH4 6) as follows.
4LiH þ Si , Li4 Si þ 2H2
*Graduate
H ð298 KÞ ¼ 120 kJ/mol-H2
ð1Þ
2MgH2 þ Si , Mg2 Si þ 2H2
H ð298 KÞ ¼ 38:9 kJ/mol-H2
ð2Þ
LiH þ LiOH ! Li2 O þ H2
H ð298 KÞ ¼ 23:3 kJ/mol-H2
ð3Þ
Student, Tohoku University
H ð298 KÞ ¼ 48:1 kJ/mol-H2
ð4Þ
2LiBH4 þ MgH2 , 2LiH þ MgB2 þ 4H2
H ð298 KÞ ¼ 25 kJ/mol-H2
ð5Þ
These reactions consist of hydrogen desorption exothermically which could lead to reduction of capacity of the heater
for hydrogen generation system. Some of these reactions
show reversibility of hydrogenation and dehydrogenation.
From a point of view for reduction of heat for hydrogen
desorption, the hydrolysis reactions of hydrides are paid
much attentions as hydrogen source. Hydrolysis reactions
could be sorted out as two groups by means of water supply.
One is direct addition of water for hydrolysis reaction, and
the other is indirect supply of water by means of such as
decomposition and water generation, i.e. dehydration, of
metal hydroxides.
Hydrolysis reactions of hydrides, aluminum hydrides and
borohydrides of Na, Li, Ca and Mg for hydrogen desorption
were reported by V. C. Y. Kong et al.7)
MHx þ xH2 O ! M(OH)x þ xH2
ð6Þ
Where M represents a metal of valence x. Hydrolysis reaction
produces hydrogen gas and alkaline hydroxide as water
solutions. The hydrolysis reaction rates of some hydrides are
high. In addition, it is possible to steady constant reaction rate
during a wide range of yield, and overall yields of hydrolysis
reaction over 96% and 90% were reported for CaH2 and LiH,
respectively.7)
Some catalysts for hydrolysis reaction of NaBH4 8,9) and
NH3 BH3 10) were also investigated in order to improve
reaction rate, although, the disadvantage of this reaction is
producing high pH solution as byproduct.
The hydride-hydroxide reaction could take place without
addition of water, and produces hydrogen gas and solid
oxide.
MH þ MOH ! MO þ H2
ð7Þ
1856
M. Kawakami et al.
The byproducts of hydride-hydroxide reactions are easier to
be handled than those of the hydrolysis reactions.
As for hydrogen generation, solid-state reactions of ionic
hydrides with alkaline hydroxides was reported by J. J. Vajo
et al.5) For example, the reaction of lithium hydride with
lithium hydroxide is described as follows.
The composite of LiH-LiOH decomposed into Li2 O and H2
with exothermic reaction. The amount of H2 generation
through this reaction is measured to be 5.73 mass% from R.T.
up to 250 C. J. J. Vajo et al. suggested that eq. (3) consist
of the two-step solid-gas reactions:
2LiOH ! Li2 O þ H2 O(g)
H ð298 KÞ ¼ 129.4 kJ/mol-H2 O
Heat Flow, Q / Wg-1
ð3Þ
2.
The starting materials, LiH (95% purity), LiOH (98%
purity) were purchased from ALDRICH. Before preparation
of samples, XRD patterns of these starting materials were
taken, and no phase was observed other than that of starting
material, LiH or LiOH respectively. The hydride-hydroxide
composites, LiH-x mol%LiOH (x ¼ 30{60) were prepared
using hand milling in an argon filled glove box with the
milling time of 30 min.
The decomposition and reaction temperatures and heats of
reactions of the starting materials and LiH-LiOH composites
were measured by using differential scanning calorimetry in
a glove box filled with argon. The samples were heated up to
420–500 C with a rate of 5 C/min under flowing argon. The
hydrogen desorption temperatures of LiH-LiOH composites
were measured by thermal desorption mass spectroscopy
combined with thermogravimetry and differential thermal
analysis heating up to 550 C with a rate of 5 C/min under
flowing helium. During sample preparation for TDS, samples
were exposed to air for several tens of minutes. The phase
identification was carried out by X-ray diffraction.
248 ºC
19.8 kJ/mol
100
200
300
400
500
Fig. 1 DSC curve of LiH-50 mol%LiOH composite. The sample was
heated up to 420 C with a rate of 5 C/min under flowing argon.
ð9Þ
Experimental Procedures
0
Temperature, T / ºC
ð8Þ
Equation (8) seems to dominate the hydrogen desorption
temperature of the composite.
On the other hand, J. M. Kiat et al. investigated the
dehydration of LiOH into Li2 O.11) Equation (8) occurs
above 350 C under the partial pressure of water vapour:
PH2 O ¼ 0{150 mmHg. With this result, eq. (3) below 350 C
should be considered to be a solid-solid reaction. The
difference between working temperature of LiH-LiOH
composite reported by Vajo and dehydration temperature of
LiOH reported by Kiat is large. So there could exist some
probability of dependency of working temperature of LiHLiOH composite on LiOH contents. Also, other properties
might be changed with changing of LiOH ratio in LiH-LiOH
composite.
The purpose of this study is to investigate the effect of
variation LiOH content for hydrogen desorption properties
of LiH-(30–60) mol%LiOH composites.
412 ºC
5.6 kJ/mol
2
-2
0
2LiH þ H2 O(g) ! Li2 O þ 2H2
H ð298 KÞ ¼ 87:9 kJ/mol-H2
5 ºC/min
Ar flow
LiH-50 mol%LiOH
4
Intensity (arbitrary units)
LiH þ LiOH ! Li2 O þ H2
H ð298 KÞ ¼ 23:3 kJ/mol-H2
6
LiH
LiOH
Li2O
LiH-50 mol%LiOH
after DSC measurement
LiOH
Li2O
prepared composite
LiH
LiOH
10º
20º
30º
40º
50º
60º
70º
80º
90º
2θ (Cu-Kα)
Fig. 2 XRD patterns of LiH-50 mol%LiOH composite of as-prepared and
after DSC measurement up to 420 C.
3.
Results and Discussions
Figure 1 shows DSC curve of LiH-50 mol%LiOH. An
exothermic reaction occurred in the temperature range of
250–410 C. The enthalpy change of this reaction is calculated as 19:8 kJ/mol-H2 . This value is in good agreement
with that of eq. (3). Around 400 to 450 C, a couple of
endothermic and exothermic reactions were observed on
heating and cooling processes, respectively. Heat flow peaks
like these imply occurrences of a phase transformation and in
this sample, which is derived from that of LiOH.2) It is
notable that onset temperature of exothermic reaction is quite
lowered than melting point (689 C) of LiH.
Figure 2 shows XRD patterns of LiH-50 mol%LiOH, of
as-prepared and after DSC measurement up to 420 C. Li2 O
and LiOH were observed in the sample after DSC measurement. The exothermic reaction in Fig. 1 around 250–400 C
could be attributed to the occurence of reaction (3).
The effect of variation of LiOH content on hydrogen
desorption properties of LiH-(0–100) mol%LiOH composites
was investigated. Figure 3 shows TDS curves of LiH(0–100) mol%LiOH composites, in which (a) mass#2 and
(b) #18 correspond to emission of hydrogen gas and water,
respectively. For pure LiH, small amount of hydrogen
desorption was obserbed to be around 100–150 C. It may
be able to attribute the hydrogen desorption derived from
LiH to contact with air. For the sample of LiH-(30–60)
mol%LiOH, the hydrogen desorption started around 200–
300 C. J. J. Vajo reported that hydrogen desorption starts at
Optimum Hydrogen Desorption Properties in LiH-LiOH Composites
(a) LiH-x mol%LiOH
(x = 0-100)
mass#2: H2
1857
(b) LiH-x mol%LiOH
(x = 0-100)
mass#18:
H2O
x = 60
Ion Current (arbitrary units)
Ion Current (arbitrary units)
x = 100
292 ºC
278 ºC
x = 50
x = 45
x = 40
x = 30
x=0
0
100
285 ºC
263 ºC
262 ºC
422 ºC
x = 60
394 ºC
x = 50
378 ºC
x = 45
374 ºC
x = 40
340 ºC
x = 30
x=0
105 ºC
200
x = 100
300
400
500
600
0
100
Temperature, T / ºC
200
300
400
500
600
Temperature, T / ºC
Fig. 3 TDS curves of LiH-(0–100) mol%LiOH composites. The samples were heated up to 550 C with a rate of 5 C/min under flowing
helium.
10
Weight Loss, W (mass%)
50–60 C for mechanically milled LiH-50 mol%LiOH without TiCl3 added as a catalyst. The differance of hydrogen
desorption temperature was caused by the sample preparation. In general, the mechanical milling refines the particle
size of the sample, so that it is possible to raise reaction rate.
LiOH decomposed and generated of water at 422 C. This
result conformed with that reported by J. M. Kiat.11) For the
sample of LiH-60 mol%LiOH, intensity of water generation
was stronger than any other composite. From the view point
of hydrogen generation materials or hydrolysis, detection of
water during reaction means some amount of water was
unutilized for the hydrogen generation, or in other word, it
could be regarded as excessive addition of water beyond
optimum content (in this case, excess of hydroxide) and
which could lead to deteriorating of gravimetrical efficiency
as hydrogen generating materials.
Meanwhile, amount of water genaration from LiH-LiOH
composites decreased with decreasing of LiOH contents
significantly from LiH-(30–50) mol%LiOH.
Figure 4 shows TG curves of LiH-(30–60) mol%LiOH.
Measured weight loss after TG measurement and calculated
hydrogen content of LiH-(30–60) mol%LiOH are summarized in Table 1. The weight loss of sample after TG measurement increased with increasing of the LiOH content. In the
composites of LiH-(50 and 60) mol%LiOH, measured weight
losses were larger than the theoretical hydrogen contents of
these compositions. This result could be due to detection of
water from LiH-LiOH composites caused by excess of LiOH
content since molecular weight of water is nine times as
heavy as that of hydrogen.
Figure 5 shows XRD patterns of LiH-(30–60) mol%LiOH
composites after DSC measurements. From all samples, Li2 O
5
LiH-x mol%LiOH
(x = 30-60)
0
x = 30
x = 40
-5
x = 45
x = 50
-10
-15
-20
0
x = 60
x = 50
x = 45
x = 40
x = 30
100
x = 60
200
300
400
500
600
Temperature, T / ºC
Fig. 4 TG curves of LiH-(30–60) mol%LiOH composites. The samples
were heated up to 550 C with a rate of 5 C/min under flowing helium.
Table 1 Weight losses after TG measurement and calcurated hydrogen
contents of LiH-(30–60) mol%LiOH composites.
Composition of
LiH-x mol%LiOH (x ¼ 30{60)
Weight loss
(mass%)
Hydrogen content
(mass%)
x ¼ 60
16.91
5.74
50
9.21
6.32
45
5.67
6.65
40
4.24
7.02
30
3.16
7.91
was observed after DSC measurement. Existance of Li2 O in
LiH-LiOH composites after DSC measurement was attributed to an exothermic reaction in the temperature range
of 250–410 C of a solid-solid reaction between LiH and
LiOH.
1858
M. Kawakami et al.
300
Hydrogen desorption temperature, T / ºC
LiH
LiOH
Li2O
LiH-x mol%LiOH
(x = 30-60)
Intensity (arbitrary units)
after DSC measurement
LiOH
x = 60
Li2O
LiOH
x = 50
Li2O
x = 45
Li2O
280
270
260
250
30
40
50
60
Content of LiOH (mol%)
x = 40
Li2O
LiH
10 º
20 º
30 º
40 º
50 º
60 º
70 º
Fig. 6 Hydrogen desorption temperatures of LiH-LiOH composites as a
function of content of LiOH.
x = 30
Li2O
80 º
90º
2θ (Cu-Kα)
Fig. 5 XRD patterns of LiH-(30–60) mol%LiOH composites after DSC
measurement up to 420 C.
In the composites of LiH-(50 and 60) mol%LiOH or LiH30 mol%LiOH, unreacted LiOH or LiH was also observed.
These results indicate that these composite were unoptimum
composition for LiH-LiOH composite. Excess amount of LiH
(or LiOH) could increase amount of unreacted LiH (or LiOH)
after hydrogen generation and deteriorate hydrogen generation gravimetrical efficiency.
In the composites of LiH-x mol%LiOH (x ¼ 40 and 45),
only Li2 O was observed. It can be said that LiH and LiOH
reacted well at these composition ratios. But, detection of
LiH is relatively difficult, so there exists some possibility that
these samples (x ¼ 40 and 45) still contained LiH after DSC
measurement.
In this study, judging from the results of XRD, TDS and
TG, suitable content of LiH-x mol%LiOH composites for
hydrogen desorption is around x ¼ 30{40. As mentioned
above, hydrogen desorption temperature of LiH-LiOH
composites are lowered significantly than that of LiH without
LiOH. In the next, effects of LiOH contents on reduction of
onset temperature of hydrogen desorption are studied.
Figure 6 shows hydrogen desorption temperatures of LiHLiOH composites as a function of the molar content of LiOH.
Hydrogen desorption temperature of LiH-LiOH composite
decreased with decreasing of the molar content of LiOH in
LiH-LiOH composites.
In general, solid-solid reaction proceeds from the surface
of each particles. The mechanical milling and addition of a
catalyst were offen used in order to enhance these surface
states. In the case of the LiH-LiOH composite, dependence of
the hydrogen desorption temperature on the content of LiOH
was observed in this study. The contact area between LiH and
LiOH might been increases with increasing the contents of
LiH.
4.
290
LiH-x mol%LiOH
(x = 30-60)
Summary
The effect of variation of LiOH content on hydrogen
desorption properties of LiH-(30–60) mol%LiOH composites
was investigated. For the studied compositions of LiH-
(30–60) mol%LiOH, the addition of LiOH destabilized
LiH to generate hydrogen below 300 C. Especially, LiH30 mol%LiOH composite shows the lowest hydrogen desorption temperature at 262 C in this study. The hydrogen
desorption temperature of these composites decreases with
decreasing the LiOH content.
From results of TDS measurements, intensity of water of
LiH-60 mol%LiOH sample was stronger than any other
sample. Intensity of water decreased with decreasing the ratio
of LiOH. No peak of generation of water was observed from
LiH-30 mol%LiOH sample.
From XRD patterens of samples after DSC measurement,
samples of LiH-(40 and 45) mol%LiOH were confirmed as
consisting of Li2 O. Meanwhile, unreacted starting material of
LiH or LiOH was observed besides Li2 O in some samples
after DSC measurement, the former was confirmed in the
sample of LiH-30 mol%LiOH, the latter was confirmed in the
samples of LiH-(50 and 60) mol%LiOH, respectively.
Judging from the results of XRD, TDS and TG, LiH-(30
and 40) mol%LiOH is a suitable composition with regard to
hydrogen desorption properties. This result could be attributed to enhancement of the contact area between LiH and
LiOH.
REFERENCES
1) C. E. Messer, E. B. Damon, P. C. Maybury, J. Mellor and R. A. Seales:
J. Phys. Chem. 62 (1958) 220–222.
2) F. H. Ellinger, C. E. Holley, Jr., B. B. McInteer, D. Pavone, R. M.
Potter, E. Staritzky and W. H. Zachariase: J. Am. Chem. Soc. 77 (2000)
157–166.
3) J. F. Stampfer, C. E. Holley and J. F. Suttle: J. Am. Chem. Soc. 82
(1960) 3504–3508.
4) J. J. Vajo, F. Mertens, C. C. Ahn, R. C. Bowman, Jr. and B. Fultz: J.
Phys. Chem. B 108 (2004) 13977–13983.
5) J. J. Vajo, S. L. Skeith, F. Mertens and S. W. Jorgensen: J. Alloy.
Compd. 390 (2005) 55–61.
6) J. J. Vajo, S. L. Skeith and F. Mertens: J. Phys. Chem. B 109 (2005)
3719–3722.
7) V. C. Y. Kong, F. R. Foulkes, D. W. Kirk and J. T. Hinatsu: Int. J.
Hydrogen Energ. 24 (1999) 665–675.
8) S. C. Amendola, S. L. Sharp-Goldman, M. S. Janjua, N. C. Spencer,
M. T. Kelly, P. J. Petillo and M. Binder: Int. J. Hydrogen Energ. 25
(2000) 969–975.
9) S. Suda, Y.-M. Sun, B.-H. Liu, Y. Zhou, S. Morimitsu, K. Arai, M.
Uchida, Y. Candra and Z.-P. Li: Appl. Phys. A 72 (2001) 209–212.
10) M. Chandra and Q. Xi: J. Power Sources 156 (2006) 190–194.
11) J. M. Kiat, G. Boemare, B. Rieu and D. Aymes: Solid State Commun.
108 (1998) 241–245.