Industrial production of MgH2 and its application

Article abstract of DOI:10.1016/j.jallcom.2010.11.047

A process for the industrial production of magnesium hydride (MgH ₂) based on a thermal equilibrium method and its application to portable hydrogen sources is proposed. Mg powders and tablets compressed with mechanically ground Mg ribbons were successfully hydrogenated in a 50-kg-batch furnace. The resultant MgH₂ showed a hydrogen yield of around 96% with good reproducibility. The compression ratio of Mg tablets was found to be an important factor in the hydrogenation yield. A hydrolysis technique using citric acid as an additive was employed to generate hydrogen under 373 K. Increasing the concentration of citric acid and the temperature accelerated the hydrolysis reactivity. A hydrogen reactor utilizing hydraulic head pressure was developed. It generated hydrogen continuously for 1 h at a flow rate of 100 ml min^-1 with hydrolysis of 5 g of tablet-form MgH₂. The conversion yield was around 70%, regardless of the flow rate.

Full text of DOI:10.1016/j.jallcom.2010.11.047

Journal of Alloys and Compounds 509S (2011) S650–S653
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Industrial production of MgH2 and its application
H. Uesugia,∗, T. Sugiyamaa, H. Niia, T. Itoa, I. Nakatsugawab
a BIOCOKE Lab. Ltd., Japan
b Fuji Kogyo Co. Ltd., Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 July 2010
Received in revised form 18 October 2010
Accepted 8 November 2010
Available online 12 November 2010
A process for the industrial production of magnesium hydride (MgH2) based on a thermal equilibrium
method and its application to portable hydrogen sources is proposed. Mg powders and tablets compressed
with mechanically ground Mg ribbons were successfully hydrogenated in a 50-kg-batch furnace. The
resultant MgH2 showed a hydrogen yield of around 96% with good reproducibility. The compression ratio
of Mg tablets was found to be an important factor in the hydrogenation yield. A hydrolysis technique using
citric acid as an additive was employed to generate hydrogen under 373 K. Increasing the concentration
of citric acid and the temperature accelerated the hydrolysis reactivity. A hydrogen reactor utilizing
hydraulic head pressure was developed. It generated hydrogen continuously for 1 h at a flow rate of
100 ml min−1 with hydrolysis of 5 g of tablet-form MgH2. The conversion yield was around 70%, regardless
of the flow rate.
Keywords:
Magnesium hydride
Hydrogen generation
Hydrolysis
Fuel cell
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
article, the scalability of this process is verified with a 50-kg-batch
furnace using Mg tablets and powder as samples.
Mg-based hydrides are promising candidates for hydrogen
storage applications because of their high gravimetric/volumetric
hydrogen storage capacity and the low cost of the raw material
[1–7]. On the other hand, high absorption/desorption temperature
and sluggish reaction kinetics require complex procedures such as
tailoring Mg alloys and compounds or by producing nanometer-
size grains through reactive ball milling [1,2]. Some procedures
have been proven to be effective in reducing the absorption tem-
perature while the desorption temperature remains high. Few
companies produce MgH2 on a commercial basis, apart from chem-
ical reagent suppliers. This can be partly explained by the complex
activation treatment required for hydrogenation [3] and the careful
handling of Mg powder, which poses the potential hazard of dust
explosion [4]. Akiyama et al. [5] have proposed hydriding chemi-
cal vapor deposition directly synthesizing single-crystal nano-sized
MgH2 fibers via gas phase reaction. The resultant MgH2 showed a
plateau of 7.6 wt% at 578 K in the P–C–T curve without any activa-
tion treatment. In collaboration with Akiyama, the present authors
have developed an industrial process for producing MgH2 by uti-
lizing thermal equilibrium between {Mg + H2} and {MgH2} [6,7].
Mg powders were hydrogenated at controlled temperature and
hydrogen pressure in a 5-kg-batch furnace. Twenty-four consec-
utive trials provided an average hydrogenation yield of 96%. In this
The hydrolysis of alkali, alkali-earth metal hydrides, or boro-
hydrides has recently attracted attention as a method of hydrogen
production for small to medium-sized polymer electrolyte fuel cells
(PEFCs). Hydrolysis with LiH or CaH2 is violent and often becomes
uncontrollable, while NaBH4 [8] shows more controllable reactiv-
ity in the presence of a catalyst. In the case of MgH2, the reaction
can be expressed as
−1
MgH2 + 2H2O → Mg(OH)2 + 2H2, ꢀH = −268 kJ mol-MgH2
(1)
Several hydrolysis studies of MgH2 have been conducted using
ball-milled MgH2 powder supplemented with graphite [9] or rare-
earth metal [10]. The compounds generate hydrogen in pure water
at room temperature. There have been no studies on how to control
such high reactivity for practical applications. In this study, hydrol-
ysis was investigated with tablet-type MgH2 in citric acid. Citric
acid is a weak organic acid that is extensively used in the food and
beverage industry. Its solution can be easily prepared without the
need for handling the dangerous concentrated acid. After charac-
terizing the process parameters, a portable reactor that generates
hydrogen at a constant rate was developed.
2. Experimental
Some 99.8%-purity Mg powder (Chuo-kosan, Japan) with a 200 mesh (<75 ␮m)
was used. Tablet-form Mg was prepared by grinding a 99.8% Mg ingot (Chuo-kosan,
Japan) into ribbons of thickness less than 150 ␮m. These ribbons were then com-
pressed in a Ø10 mm steel mold. The compression ratio was controlled by applying
pressure. The apparent density was calculated by the weight and volume.
∗
Corresponding author. Tel.: +81 368092082;fax: +81 368092083.
E-mail addresses: uesugi-h@biocokelab.com, uesugi-bcg@kind.ocn.ne.jp
(H. Uesugi).
0925-8388/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2010.11.047

H. Uesugi et al. / Journal of Alloys and Compounds 509S (2011) S650–S653
S651
100
80
60
40
20
30 h
45 h
0.3
0.6
0.9
1.2
1.5
Density (gcm 3)
Fig. 2. Relation between density of Mg tablet and hydrogenation yield. Hydrogena-
tion was carried out for 30 h and 45 h.
minimum of 93% and averaged 97%. X-ray powder diffraction was
also measured to detect only those peaks related to Mg and MgH2.
The presence of oxide thus seems negligible. The variation from
batch to batch was slightly higher than the previous trials with
a 5-kg-batch furnace. Fig. 2 shows the relationship between the
density of Mg tablet and hydrogenation yield for various hydro-
genation periods. Longer hydrogenation gave better yields. The
positive effect of tablet density on hydrogenation yield is evident.
Generally, hydrogen absorption in metal improves remarkably with
plastic deformation, which introduces lattice defects to function
as trap sites for hydrogen. In hydrogen storage studies, accumu-
lative roll bonding or severe plastic deformation have been found
to improve absorption characteristics such as ease of activation or
higher capacity [11]. The pressure applied to fabricate Mg tablets
certainly induces plastic deformation inside Mg ribbons and resul-
Fig. 1. MgH2–hydrogen reactor (MHR) for hydrolysis of MgH2 tablet in citric acid.
The hydrogenation of Mg samples was conducted based on a thermal equi-
librium method [6,7]. Powder and tablet Mg were placed on trays inside a 50 kg
hydrogenation furnace. After purging with Ar, the furnace was filled with 99.9999%
hydrogen with a pressure of less than 2 MPa. The temperature of the process was
controlled so as to remain near the boundary between the regions {MgH2} and
{Mg + H2} in the thermal equilibrium diagram, and maintained for 30–45 h. The
hydrogen consumption behavior was monitored by a pressure gauge. Afterward,
the furnace was gradually cooled down to room temperature. The hydrogenation
yield (%) was calculated via the weight difference of Mg samples before/after the
process.
One MgH2 tablet (∼1.7 g) was placed into a 200 ml Pyrex glass flask. Some 55 ml
of 0.52–1.04 mol l−1 citric acid solution prepared by reagent grade chemicals and
de-ionized water was added to react with the tablet. The experiment was carried
out under atmospheric pressure and a temperature of 299–343 K in a water bath. In
addition, 200 mg MgH2 powder in 10 ml citric acid was studied. The hydrogen that
emerged from the flask was collected in an inverted mess cylinder to measure the
quantity. The amount of hydrogen produced was expressed by conversion yield (%),
which is defined as the volume of produced hydrogen over the theoretical volume of
hydrogen, and by hydrogen generation rate (ml min−1 g MgH2−1), and all materials
were assumed to react during hydrolysis reaction (1).
A hydraulic head-type MgH2 reactor (MHR) consisting of two containers, shown
in Fig. 1, was developed. Three pieces of MgH2 tablets (∼5 g) were placed in the
lower container, and 100 ml of 1.04 mol l−1 citric acid was poured into the upper
container. By manipulating the pressure in the upper and lower containers, a suit-
able amount of citric acid descended to continue hydrolysis in the lower container.
The initial pressure was set to 0.06–0.08 MPa. The hydrogen generation rate was
regulated to 50, 100, and 150 ml min−1 by a flow meter equipped with a control
valve. The time taken for the pressure in the lower container to decrease to below
0.02 MPa was monitored. For reference, hydrolysis with MHR under atmospheric
pressure without flow control (open system, the same experiment as with the flask)
was also conducted. The experiment was repeated at least three times to confirm
reproducibility.
tant lattice defects. Considering the Mg density of 1.74 g cm−3
,
Mg tablets with apparent density of 1.2–1.5 g cm−3 would provide
the optimum condition for hydrogenation by creating numerous
lattice defects while maintaining sufficient voids for hydrogen dif-
fusion. Hereafter, MgH2 tablets with 95% hydrogenation yield were
employed.
3.2. Hydrolysis of MgH2 with citric acid
Fig. 3 shows an example of the hydrolysis behavior of MgH2
in citric acid. The conversion yield follows a parabolic curve; the
increase is reduced at longer hydrolysis periods. Some researchers
have attempted to analyze the trajectory by curve fitting based
on the Avrami–Erofeev reaction [9,10]. The hydrogen generation
rate was 161 ml min−1 g−1 initially and decreased to 3 ml min−1 g−1
after 60 min, 1.9% of the initial rate.
Table 2 summarizes the effect of process parameters on the time
required to attain a conversion yield of 80% after 60 min, and an
average generation rate within 60 min. For reference, the results of
MgH2 powder are also mentioned. Hydrolysis with MgH2 powder
completed within a minute, and the yield reached 100%. On the
other hand, MgH2 tablet reaction required a much longer period.
The average hydrogen generation rate was proportional to the con-
centration of citric acid. Increasing the preset temperature was also
effective in improving hydrolysis kinetics. Note that the hydrolysis
reaction (1) is highly exothermic; the actual solution temperature
would not be the same as that of the water bath once the reac-
tion had started. Nevertheless, the results suggest that the heat is
effective in continuing the reaction. The temperature dependence
3. Results and discussions
3.1. Hydrogenation of Mg in 50-kg-batch furnace
Table 1 shows the result of four batch trials with a load of up
to 23 kg of powder and tablet Mg. The hydrogenation yield was a

S652
H. Uesugi et al. / Journal of Alloys and Compounds 509S (2011) S650–S653
Table 1
Hydrogenation yield of Mg powder and tablet at 50 kg batch furnace. Hydrogenation was carried out for 45 h.
RUN
Total Mg (kg)
Mg powder
Weight (kg)
Mg tablet
Hydrogenation (%)
Weight (kg)
Hydrogenation (%)
1
2
3
4
8
15
16
23
8
15
16
17
97.6
93.2
100.0
93.1
0.1
0.3
0.2
5.6
98.4
95.3
99.0
97.9
97.7
Average
96.0
Table 2
Effect of process parameters on hydrolysis of MgH2.
MgH2
Citric acid (mol l−1
)
Temperature (K)
t at conversion yield = 80% (min)
Conversion yield after t = 60 min (%)
Average H2 rate (ml min−1 g−1
)
Powder
0.26
0.52
299
299
0.6
0.1
100.0
99.6
2.87 × 103
9.06 × 103
Tablet
0.52
0.78
1.04
0.78
0.78
0.78
0.78
299
299
299
313
323
333
343
60<
60<
35.3
60<
60<
53.9
70.0
86.0
75.1
78.9
83.2
92.5
16.2
21.1
24.3
22.7
23.4
24.7
27.6
48.0
25.4
80
70
140
120
100
of the solubility of Mg byproduct would undoubtedly influence this
behavior.
3.3. Hydrogen generation with MHR using MgH2 tablet
Duration
Conversion yield
As shown in Table 2, MgH2 powder is more reactive than in
tablet form. However, from the viewpoint of reaction control, tablet
MgH2 is advantageous. With our designed MHR, the height of the
solution level in the lower container changes according to the
hydraulic head pressure of the higher and lower containers. If the
hydrolysis reaction proceeds a little further, the pressure increases
to push the solution to the upper container. As a result, the surface
area in contact with the solution drops, and the reaction ceases.
In addition, tablets prepared by Mg ribbons greatly reduce the risk
of dust explosion during handling and associated cost. Thus, tablet
MgH2 was used in MHR studies.
80
60
40
20
0
60
50
40
50
100
150
Open
(2hrs)
Fig. 4 shows of the time taken for hydrogengenerationwith MHR
under various flow rates. The average conversion yield was also
plotted. During operation, hydrogen flow was almost steady once
it was set, while the pressure decreased gradually with time. When
the pressure in the lower container decreased to 0.04 MPa, flow
rate adjustment with a control valve became necessary. In general,
the MHR supplied a constant flow of hydrogen autonomously until
Flow rate (ml min-1)
Fig. 4. Duration period and conversion yield obtained at MHR under constant flow
rates. Hydrolysis was carried out with 5 g MgH2 tablet in 100 ml of 1.04 mol l−1 citric
acid, under flow rate control.
90% of the total duration. The obtained conversion yield of around
70% was similar under all conditions and was close to the value
obtained with the open system. These findings indicate that the
MHR can work as a hydrogen generator with simple control. The
present capacity of hydrogen production is sufficient for PEFCs of
5–10 W. Since it is possible to scale up the MHR system without
major modifications, applications to small to medium-sized PEFCs
are promising.
80
60
40
20
0
200
150
100
50
conversion yield
flow rate
Further studies on hydrolysis conditions with fewer addi-
tives, and the recycling or regeneration of byproducts [7], are in
progress.
4. Conclusion
Tablet and powder Mg were hydrogenated in a 50-kg-batch
furnace. The resultant MgH2 had a hydrogenation yield of 93% or
higher. Hydrogenation was influenced by the pressure applied for
preparing the Mg tablet, which induced mechanical deformation in
the Mg ribbon to create trap sites for hydrogen. A hydrolysis tech-
nique using citric acid was employed to generate hydrogen under
0
0
10
20
30
40
50
60
t (min)
Fig. 3. Change of conversion yield and hydrogen flow rate with time. Hydrolysis
was carried out with 1 g MgH2 tablet in 33 ml of 1.04 mol l−1 citric acid, T = 299 K,
under open condition.

H. Uesugi et al. / Journal of Alloys and Compounds 509S (2011) S650–S653
S653
373 K. A high concentration of citric acid and elevated temperature
accelerated the hydrolysis reactivity.
[4] Kyoto University and Material Process Technology Center, “Magnesium Forged
Material Technology Development Project,” First Committee document No. 7-3
NEDO Japan, 2008.
[5] I. Saita, T. Toshima, S. Tanda, T. Akiyama, J. Alloys Compd. 446–447 (2007)
80–83.
[6] T. Akiyama, X. Zhang, M. Sato, H. Hayashi, T. Hiraki, I. Saita, and H. Uesugi, Jpn.
Patent 2008-44832.
[7] H. Uesugi, T. Sugiyama, I. Nakatsugawa, T. Ito, Proc. 67th Ann. World Mg. Conf.,
Int. Magnesium Assoc., Hong Kong, May 2010, pp. 49–59.
[8] S.C. Amendola, S.L. Sharp-Goldman, M.S. Janjua, M.T. Kelly, P.J. Petillo, M. Binder,
J. Power Sources 85 (2000) 186–189.
A hydrogen reactor based on the hydrolysis of MgH2 tablet
was developed. The reactor supplied a constant hydrogen flow
autonomously for 90% of the total duration. The conversion yield
was around 70%, regardless of the flow rate. Thus, application of this
system as a portable hydrogen generator for PEFCs is anticipated.
References
[9] R.V. Lukashev, N.A. Yakovleva, S.N. Klyamkin, B.P. Tarasov, Russ. J. Inorg. Chem.
53 (2008) 343–349.
[1] X.D. Yao, G.Q. Lu, Chin. Sci. Bull. 53 (2008) 2421–2431.
[2] M. Jehan, Proc. 66th Ann. World Magnesium Conf., Int. Magnesium Assoc., San
Francisco, June 2009.
[10] L.Z. Ouyang, Y.J. Xu, H.W. Dong, L.X. Sun, M. Zhu, Int. J. Hydrogen Energy 34
(2009) 9671–9676.
[11] G.F. Lima, A.M. Jorge Jr., D.R. Leiva, C.S. Kiminami, C. Bolfarini, W.J. Botta, J. Phys.
Conf. Ser. 144 (2009) 012015.
[3] I. Saita, T. Akiyama, J. Chem. Eng. Jpn. 39 (2006) 525–530.