ww.sciencedirect.com

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e8

Available online at w

ScienceDirect

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

Experimental study of metal hydride-basedhydrogen storage tank at constant supply pressure

A. Souahlia, H. Dhaou*, S. Mellouli, F. Askri, A. Jemni, S. Ben Nasrallah

Laboratoire d’Etudes des Systemes Thermiques et Energetique, Rue Ibn Eljazzar,

Ecole Nationale d’Ingenieurs de Monastir, University of Monastir, Monastir 5019, Tunisia

a r t i c l e i n f o

Article history:

Received 11 December 2013

Received in revised form

14 February 2014

Accepted 19 February 2014

Available online xxx

Keywords:

Hydrogen

Metal hydride

Supply pressure

Cooling temperature

* Corresponding author. Tel./fax: þ216 73501E-mail address: dhaou_2000tn@yahoo.fr

Please cite this article in press as: Souaconstant supply pressure, International J

http://dx.doi.org/10.1016/j.ijhydene.2014.02.10360-3199/Copyright ª 2014, Hydrogen Ener

a b s t r a c t

Metal hydride-based hydrogen storage tank is tested using 1 kg of AB5 alloy, namely LaNi5.

The hydrogen tank is of annular cylindrical with inner and outer heat exchangers. The

inner one is a finned spiral heat exchanger and the outer one is a conventional jacket.

Performance (storage capacity and storage time) studies are carried out by varying the

supply pressure and the cooling temperature of the hydride bed. At any given cooling

temperature, hydrogen storage rate is found to increase with supply pressure. Cooling

temperature is found to have a significant effect on hydrogen storage capacity at lower

supply pressures.

Copyright ª 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

Introduction

Since environmental pollution proceeds from the large

amount of fossil fuel energy consumption, new clean energy

sources have been developed as alternatives. In addition,

hydrogen is one of the most clean energy carriers. However,

hydrogen storage and transportation become problematic.

Among different hydrogen storage means (compressed gas,

liquid hydrogen) reversiblemetal hydrides are considered as a

safe and volume efficient hydrogen storage medium under

low pressure conditions. Metal hydrides are generally used as

packed beds. Themetal hydride formation is closely related to

the hydrogen pressure and bed temperature. Also, hydrogen

absorption/desorption is an exothermic/endothermic reac-

tion. Those phenomena and many others are complicatedly

coupled together, so it requires a lot of experimental and

597.(H. Dhaou).

hlia A, et al., Experimenournal of Hydrogen Ene

21gy Publications, LLC. Publ

theoretical works to optimize hydrogen metal hydride

reservoir.

Suda and Kobayashi [1] performed a series of experimental

studies on the hydriding and dehydriding kinetics for several

alloys and their binary mixtures under isothermal conditions.

They concluded that the mixing of hydriding materials offers

several technical advantages such as controlling the reaction

rate and the kinetic properties by varying the mixing ratio.

Miyamoto et al. [2] investigated the reaction kinetics for

LaNi5eH2 system under constant hydrogen pressure, and

proposed the chemical reaction rate models. Goodell and

Rudman [3] measured the intrinsic reaction rates for the

hydriding and dehydriding of LaNi5 over a wide range of

pressures in the temperature range 60e65 �C. Approximately

isothermal conditions were maintained by a thermal ballast

technique. Later numerical models were developed for

assessing the transient heat and mass transfer within metal

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

ished by Elsevier Ltd. All rights reserved.

Fig. 1 e Schematic of the hydride tank.

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 x x x ( 2 0 1 4 ) 1e82

hydride beds [4e7]. It was found that the heat transfer plays a

determining factor in enhancing the performance of storage

tanks. For the sake of simplicity, these models neglected the

convective effects of hydrogen gas on the heat and mass

transfer, which however may be influential at high pressure

cases.

From previous literature, the chemical reactions in metal

beds may be simplified as a two-dimensional problem. Jemni

and Ben Nasrallah developed two-dimensional heat and

mass transfer models for hydrogen adsorption and desorp-

tion respectively [8,9]. The hydrogen flow motion within the

hydride beds was described by Darcy’s law. Their results

show that the difference between the solid and hydrogen

temperatures is negligible, except for some limited areas

close to the gas outlet and tank wall. So the local thermal

equilibrium hypothesis can be used. In a subsequent study,

Ben Nasrallah and Jemni [10] further tested the hypotheses

that were usually adopted in the analysis of heat and mass

transfer in a metal hydride tank. They reached the conclu-

sion that for an LaNi5eH2 reactor, (i) solid and gas tempera-

tures can be treated as the same, (ii) the convection term is

negligible, and (iii) the effect of hydrogen concentration on

the equilibrium pressure variation is negligible in the ab-

sorption case, But in the desorption case this effect cannot be

neglected in the determination of the temperature distribu-

tion in the reactor. The simplified numerical simulations

were later compared with experiments and a good agreement

is obtained [11,12].

Recently, Tange et al. [13] examined the feasibility of an on-

site energy storage system using a tank packed with 50 kg of

hydride metal, and discuss the energy efficiency of the sys-

tem. This tank stores hydrogen at night and supply a fuel cells

to generate power during the day. The system also utilizes the

endothermic hydrogen desorption process for air

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

conditioning. D’Orazio et al. [14] analyzed and reported the

dynamic behavior of a metal hydride tank to propose a

method to design a multi-tank storage system. The hydride

tanks design and the heat flow are the key parameters for the

optimization of the system working condition. Nyamsi et al.

[15] conducted an analytical and numerical study to optimize

a finned tube heat exchanger considering both enhanced heat

transfer and metal hydride tank volume efficiency. It was

shown that the fin dimensions, the cooling tube diameter and

the fin length are the key parameters to reduce the thermal

resistance of the heat exchanger. The results showed that the

decreasing of the thermal resistance of 13% leads to a

decreasing in charging time of 42%. Corgnale et al. [16]

developed a scoping tool, referred to as the Acceptability En-

velope, to identify the range of chemical, physical and

geometrical parameters that allow a coupled media and

hydrogen storage system to meet technical targets. Nam et al.

[17] developed a three-dimensional model to study the

hydrogen absorption reaction, heat and mass transport phe-

nomena in metal hydride hydrogen storage tank. The simu-

lation results demonstrate that the use of higher hydrogen

supply pressure leads to not only rapid hydrogen charging

performance but also a reduction in the cooling burden of the

tank.

We studied, in previous work [18,19], the dynamic behavior

of several metal hydride-based hydrogen storage tanks using

a Sievert-type apparatus. We noticed that the reference vol-

ume variation influences the stored amount of hydrogen and

the storage time. In order to have consistentmeasurements of

a tank performances (storage time and storage capacity), the

reference volume influence should be excluded. The solution

would be to supply the tank with a constant hydrogen pres-

sure. In this paper, a test bench was built to implement this

solution. Also, a modified version of the metal hydride tank

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e8 3

studied by H. Dhaou et al. [18] was designed and its perfor-

mances were evaluated for various dynamic operating con-

ditions (hydrogen supply pressures, cooling temperature).

Experimental setup

The metal hydride tank considered in this paper is of annular

cylindrical configuration (Fig. 1). One kilogram (1 kg) of LaNi5alloy is packed in the inner volume of the tank. The purities of

the lanthanum and nickel were 99.80% and 99.97%, respec-

tively. The average particle size of the alloy was approxi-

mately 50 mm. Heat transfer fluid (water) flows through two

heat exchangers. The first is a finned spiral heat exchanger

immersed in the packed alloy. It consists of a stainless steel

tube (5 mm inner diameters and 1 mm thickness) bent into a

circular helix of 50 mm diameter and 11 turns with a pitch of

7 mm. A copper fin was inserted between each pair of turns

(Fig. 2). The second heat exchanger is implemented by

mounting a sealed stainless steel cylindrical jacket over the

tank, creating an annular space within which the heat trans-

fer fluid flows. The tank is made of a seamless stainless steel

(316-L). The upper end of the tank is flanged with the brazed

assembly of three-grounded metal-sheathed “K” type ther-

mocouples (sensitivity 0.1 �C) at different positions

((r ¼ 10 mm, z ¼ 25 mm), (r ¼ 30 mm, z ¼ 10 mm) and

(r ¼ 30 mm, z ¼ 20 mm)). Other instrumentation and sensors

are shown in Fig. 3. Piezoresistive-type pressure transducers

of range 0e100 bar (sensitivity 0.01 bar) are used formeasuring

the pressure inside the tank (P1) and the supply pressure (P2). A

hydrogen-calibrated rotameter (sensitivity 0.1 L/min) is used

for measuring the hydrogen flow rate and temperature of the

hydrogen gas. High-pressure ball valves are used in the gas

lines for controlling/diverting the gas flow. A vacuum pump is

used to evacuate the tank before hydrogen storage experi-

ments. A thermostatic water bath is used for supplying hot

and cold water (0e80 �C, accuracy � 0.1 �C) at flow rates

ranging from 3 to 13 g/s. A data logger is used to record

Fig. 2 e Schematic of the finned spiral heat exchanger.

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

temperature and pressure readings for further processing and

interpretation. Additional efforts were made to guarantee

accurate and precise measurement of various parameters

involved in the process of hydrogen storage. For example,

mechanical devices that generate vibrations (pump, cooling

device, devices equipped with fans.) are isolated from the

rest of the experimental setup, thermocouples, pressure

transducers and rotameter are calibrated before performing

the series of measurements presented in this paper.

To be able to reversibly store hydrogen, the LaNi5 alloy

must be activated. Activation consists in executing repeated

absorption/desorption cycles (25 cycles) until the maximum

hydrogen storage capacity is reached (1.4 w%). Absorption is

performed at 10 bar and 25 �C till the hydrogen flow is no

longer detected. Desorption reaction is performed at low

pressure and 70 �C till the pressure inside the tank becomes

constant and equal to 0.01 bar.

For each experiment, efforts are made to have the same

initial condition of the tank. First, the tank is evacuated down

to 0.01 bar. Then, the hydride bed is cooled by circulating the

cold fluid from bath B1. Meanwhile, the hydrogen supply

pressure and the fluid temperature of bath B2 are set. Later,

valve V1 is opened. The hydrogen storage is continued till the

hydrogen flow is no longer detected. During the storage pro-

cess, pressures, temperatures and hydrogen flow rate, are

recorded for every 1 s.

Results and discussion

In practice, a metal hydride tank would not be packed full of

metal hydrides but contain some void space called expansion

volume because metal lattices expand during absorption [20].

Consequently, the inner volume of a tank may be split into

two: a reactional volume corresponding to the volume occu-

pied by the totality of the hydride particles and a free volume

including the expansion volume and the hydride pores vol-

ume. In case of the studied tank, 85% of the inner volume is a

reactional volume, 12% is a free volume and not forgetting the

volume occupied by the internal heat exchanger (3%).

The terminology used in this paper distinguishes between

hydrogen absorption and hydrogen storage. The term “ab-

sorption” limits the discussion to the phenomenon occurring

in the reactional volume.

Figs. 4e6 show time history of hydrogen storage rate, hy-

dride bed temperature and hydrogen pressure inside the

reservoir, respectively. The hydrogen supply pressure is equal

to 6 bar and the cooling fluid temperature is equal to 20 �C.Overall, the hydrogen storage rate instantaneously reaches

a peak initially and then decreases gradually toward zero at

the end of the storage process (Fig. 4). Similarly, the bed

temperature increases sharply and then decreases gradually

tending toward the preset value of the cooling fluid temper-

ature (Fig. 5). Thus, the time history of the storage process can

be split into two stages A and B [19,21].

Over the A stage, at the experiment beginning, the

hydrogen storage rate increases rapidly toward a peak (Fig. 4).

Two major events contribute in the amplitude of the peak: A

rapid spread of hydrogen into the initially nearly empty free

volume and an impulsive absorption reaction induced by the

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

Fig. 3 e Synoptic scheme of the experimental setup.

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 x x x ( 2 0 1 4 ) 1e84

high-pressure difference between the hydrogen pressure in-

side the tank and the hydride equilibrium pressureðPH2 � PeqÞ,known as driving potential for mass transfer. It should be

pointed that in our case the effect of the first event is negli-

gible due to the small free volume, also because of the rapid

absorption reaction which takes over.

Fig. 4 e Time history of hydrogen storage rate. Accuracy

±0.7 g/min.

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

As time progresses, a fast and significant drop in hydrogen

storage rate and a rapid rise of both hydride bed temperature

and hydrogen pressure within the reservoir are recorded (Figs.

4 and 5). Due to the poor thermal conductivity of the hydride

bed, the generated heat of absorption is unable to be trans-

ferred from the bed to the cooling fluid at the initial period of

rapid absorption and hence the excess heat is stored in the

hydride bed itself, resulting in a sudden rise in bed tempera-

ture. Thus, the hydride equilibriumpressure increases and the

hydrogen storage rate decreases. The hydrogen pressure

within the tank increases to compensate for the drop of the

driving potential for mass transfer (Fig. 6).

The observations above prove that, along the A stage, the

evacuation of the hydrogen absorption heat via the heat ex-

changers of the reservoir has no influence on the storage rate

because of the rapidity of the reaction. In sum, at the begin-

ning, the hydrogen storage process is driven mainly by the

difference between the hydrogen pressure within the reser-

voir and the hydride equilibrium pressureðPH2 � PeqÞ.Over the B stage, the absorption rate slows down and de-

creases gradually toward zero (Fig. 4). The hydrogen pressure

inside the reservoir has an almost constant value close to the

supply pressure (Fig. 6). The bed temperature decreases

gradually tending toward the preset value of the cooling fluid

temperature (Fig. 5). This is due to an improvement of the heat

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

Fig. 5 e Time history of hydride arithmetic average of bed

temperature. Accuracy ±2 �C.

05101520253035404550

3 4 5 6 7 9 11

Volu

me

of st

ored

hyd

roge

n (L

)

Hydrogen supply pressure (bar)

Fig. 7 e Effect of supply pressure on the volume of

hydrogen stored in the tank. Accuracy ±3 L.

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e8 5

evacuation rate owing to a higher temperature gradient be-

tween hydride bed and coolant and a fall of heat generation.

The observations above prove that, along the B stage,

hydrogen storage is driven mainly by heat transfer rate

through reduction of hydride equilibrium pressure Peq.

Effect of supply pressure

The reservoir is allowed to store hydrogen at different supply

pressures maintaining constant temperature and flow rate of

cooling fluid, respectively, at 20 �C and 10 g/s.

Increasing supply pressure leads to a decrease in the vol-

ume of hydrogen stored in the tank (Fig. 7) and an increase in

the mass of hydrogen stored (Fig. 8). In fact, according to the

perfect gas law, the hydrogen density increases with pressure.

According to Fig. 9, hydrogen storage duration diminishes

with increasing supply pressure. For example, it takes 23 min

to store 10 g of hydrogen at 4 bar supply pressure, while it only

takes one and a half minutes at 11 bar supply pressure.

In Fig. 10, hydrogen storage rate increases with the supply

pressure. At higher supply pressures, the difference between

Fig. 6 e Time history of hydrogen pressure inside the tank.

Accuracy ±0.3 bar.

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

supply pressure and hydride equilibrium pressure is impor-

tant, which results in higher absorption rate and shorter

storage time.

Fig. 11 shows the time history of hydride bed temperature

for each hydrogen supply pressure. It’s observed that the rise

in bed temperature is higher at higher supply pressures. This

reflects not only an increase in absorption rate at higher

supply pressures but also an improvement in the distribution

of absorption rate inside the reactional volume (Larger num-

ber of hydride particles absorb hydrogen simultaneously). It’s

also observed from Fig. 11 that hydride bed cooling is even

faster than supply pressure increases. This is explained by the

fact that the heat transfer rate enhances with the increase of

temperature difference between the hydride bed and the

coolant. In addition, a higher hydrogen supply pressure leads

to a greater driving potential all along the storage process even

during impulsive stage of absorption (see Fig. 12), it follows

that hydride bed tends rapidly toward its maximum hydrogen

capacity, causing a faster fall in the rate of heat generation

and hence a rapid cooling of hydride bed.

To summarize, at low supply pressures, the pressure dif-

ference ðPH2 � PeqÞ drives the process of hydrogen storage for a

short span of time, hence most of the storage process is gov-

erned and controlled by the heat transfer rate. However, at

0

2

4

6

8

10

12

14

3 4 5 6 7 9 11

Mas

s of s

tore

d hy

drog

en (g

)

Hydrogen supply pressure (bar)

Fig. 8 e Effect of supply pressure on the mass of hydrogen

stored in the tank. Accuracy ±0.7 g.

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

0

2

4

6

8

10

12

14

0 500 1000 1500 2000 2500 3000

Mas

s of s

tore

d hy

drog

en (g

)

Time (s)

3 bar4 bar5 bar6 bar7 bar9 bar11 bar

Fig. 9 e Effect of the supply pressure on mass of hydrogen

stored in the tank. Accuracy ±0.5 g.

20

25

30

35

40

45

50

55

60

65

0 500 1000 1500 2000 2500 3000

Aver

age

bed

tem

pera

ture

(°C

)

Time (s)

3 bar4 bar5 bar6 bar7 bar9 bar11 bar

Fig. 11 e Effect of the supply pressure on the arithmetic

average of bed temperature. Accuracy ±2 �C.

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 x x x ( 2 0 1 4 ) 1e86

high supply pressures, pressure difference ðPH2 � PeqÞ and heat

transfer rate combine together leading to a faster hydrogen

storage process. In the case of studied tank, a pressure above

5 bar can be considered as a high pressure, because beyond

this pressure, the decrease in the time of hydrogen storage

becomes slower.

Effect of cooling fluid temperature

The effect of cooling fluid temperature on the hydrogen stored

mass illustrated on Fig. 13 show the expected trend that the

hydrogen storage capacity increases at lower cooling fluid

temperature ranges. At lower cooling fluid temperatures, hy-

dride equilibrium pressure is low which results in a larger

pressure difference between supply pressure and hydride

equilibrium pressure. This results in an increase in the

hydrogen storage capacity.

It’s also observed in Fig. 13 that, at high supply pressure,

the effect of cooling fluid temperature on hydrogen storage

capacity is less pronounced because the difference between

supply and equilibrium pressures is large.

0

5

10

15

20

25

0 500 1000 1500 2000 2500 3000

Stor

age

rate

(g/m

in)

Time (s)

3 bar4 bar5 bar6 bar7 bar9 bar11 bar

0

5

10

15

20

25

0 5 10 15 20

Fig. 10 e Effect of the supply pressure on hydrogen storage

rate. Accuracy ±0.7 g/min.

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

Identification

The main aim of system identification is to determine a

mathematical model of a physical/dynamic system from

experimental data. The identification procedure is summa-

rized in the following steps:

� Record the response of the system to a known excitation

(usually a step signal).

� According to the shape of the response, select a known

mathematical model (the simplest possible with minimal

parameters).

� Optimize the parameters of the mathematical model to fit

with the response.

The hydrogen tank studied in this paper is assimilated to a

system with a single input and a single output. The input is

the hydrogen pressure and the output is the stored mass of

hydrogen as mentioned in Fig. 14, where PH2 Pref, mstr and mss

are respectively inlet hydrogen pressure, reference pressure,

mass of stored hydrogen and maximum storage capacity of

hydrogen.

-1

0

1

2

3

4

5

6

7

8

9

10

0 200 400 600 800 1000Time (s)

3 bar4 bar5 bar6 bar7 bar9 bar11 bar

(bar

)

Fig. 12 e Effect of the supply pressure on driving potential

for mass transfer ðPH2LPeqÞ. Accuracy ±0.6 bar.

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

0

2

4

6

8

10

12

14

0 500 1000 1500 2000 2500

Mas

s of s

tore

d hy

drog

en (g

)

Time (s)

Fig. 13 e Effect of the thermostatic water temperature on

mass of stored hydrogen. Accuracy ±0.5 g.

Fig. 14 e Systemwith single input and output modeling the

tank.

00,10,20,30,40,50,60,70,80,91

0 500 1000 1500 2000 2500Time (s)

Response at 20°C

Model at 20 °C

Response at 40°C

Model at 40 °C

Fig. 16 e Time history of the response of the tank to a step

of hydrogen pressure at two different cooling temperatures

(20 �C and 40 �C).

i n t e rn a t i o n a l j o u rn a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e8 7

For LaNi5 mss, ¼ 14 (g of H2/kg of LaNi5) and assuming that

the hydrogen working pressure of the tank is equal to 5 bar,

then Pref¼ 5 bar. Fig. 15 shows the time history of the response

of the tank to a step of hydrogen pressure PH2 ¼ 5 bar at cooling

temperature of 20 �C.According to the shape of the response, the tank can be

considered as a first order system and the corresponding

mathematical model is the following:

mstrðtÞmss

¼ K

�1� exp

��ts

��

where K is the asymptotic limit of the response and s (s) is the

time constant of the system. In our case, K and s are estimated

by trial and error. Other mathematic models and parameters

estimation methods exist and will be further discussed in

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

0 500 1000 1500Time (s)

Response

Model

Fig. 15 e Time history of the response of the tank to a step

of hydrogen pressure.

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

future work. The mathematical model fit well with the

experimental response for K ¼ 0.91 and s ¼ 460 s (Fig. 15).

Fig. 16 shows that an increasing cooling water tempera-

ture results in different values of the model parameters:

K ¼ 0.83 and s ¼ 1150 s. We can conclude that s estimates the

speed of the hydrogen storage and K evaluates the mass of

stored hydrogen in relation to a referential value (mss). The

step response described above and the derived parameters

allow a quantitative comparison between the performances

of tanks.

Conclusion

A prototype of a metal hydride cylindrical tank is tested. This

tank is fitted with two heat exchangers for a better hydride

temperature regulation: an inner finned spiral heat exchanger

and a lateral heat exchanger. The important characteristics of

the metal hydride tank were analyzed by means of constant

hydrogen supply pressure tests. At any given cold fluid tem-

perature, the hydrogen storage time was found to be lesser at

higher supply pressure. Also it was found that the volume of

stored hydrogen decreases with increase of supply pressure.

Furthermore, at higher supply pressure, the heat transfer rate

is enhanced and the effect of the driving force for mass

transfer is stronger and lasts longer. Cooling fluid temperature

has a strong influence on hydrogen storage capacity and time,

however this influence is less pronounced for high supply

pressures. We discussed the response of the tank to a

hydrogen pressure step and tow parameters were derived K

and s to characterize quantitatively the performances of the

metal hydride hydrogen storage tank.

r e f e r e n c e s

[1] Suda S, Kobayashi N. Reaction kinetics of metal hydrides andtheir mixtures. J Less-Common Met 1980;73:119e26.

[2] Miyamoto M, Yamaji K, Nakata Y. Reaction kinetics of LaNi5.J Less-Common Met 1983;89:111e6.

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121

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 x x x ( 2 0 1 4 ) 1e88

[3] Goodell PD, Rudman PS. Hydriding and dehydriding rates ofthe LaNi5eH system. J Less-Common Met 1983;89:117e25.

[4] Mayer U, Groll M, Supper W. Heat and mass transfer in metalhydride reaction beds: experimental and theoretical results. JLess-Common Met 1987;131:235e44.

[5] Sun D-W, Deng S-J. Study of the heat and mass transfercharacteristics of metal hydride beds: a two-dimensionalmodel. J Less-Common Met 1989;155:271e9.

[6] Gopal MR, Murthy SS. Prediction of heat and mass transfer inannular cylindrical metal hydride beds. Int J HydrogenEnergy 1992;17:795e805.

[7] Jemni A, Ben Nasrallah S. Study of two-dimensional heat andmass transfer during absorption in a metalehydrogenreactor. Int J Hydrogen Energy 1995;20:43e52.

[8] Jemni A, Ben Nasrallah S. Study of two-dimensional heat andmass transfer during desorption in a metalehydrogenreactor. Int J Hydrogen Energy 1995;20:881e91.

[9] Ben Nasrallah S, Jemni A. Heat and mass transfer models inmetalehydrogen reactor. Int J Hydrogen Energy1997;22:67e76.

[10] Jemni A, Ben Nasrallah S, Lamloumi J. Experimental andtheoretical study of a metalehydrogen reactor. Int JHydrogen Energy 1999;24:631e44.

[11] Mat MD, Kaplan Y. Numerical study of hydrogen absorptionin an LmeNi5 hydride reactor. Int J Hydrogen Energy2001;26:957e63.

[12] Aldas K, Mat MD, Kaplan Y. A three-dimensionalmathematical model for absorption in a metal hydride bed.Int J Hydrogen Energy 2002;27:1049e56.

[13] Tange Manabu, Maeda Tetsuhiko, Nakano Akihiro,Ito Hiroshi, Kawakami Yoshiaki, Masuda Masao, et al.Experimental study of hydrogen storage with reactionheat recovery using metal hydride in a totalized hydrogenenergy utilization system. Int J Hydrogen Energy2011;36:11767e76.

Please cite this article in press as: Souahlia A, et al., Experimenconstant supply pressure, International Journal of Hydrogen Ene

[14] D’Orazio A, Dell’Era A, Artuso P, Pasquali M. Hydride tankstorage system dimensioning on the base of their dynamicbehaviour. Int J Hydrogen Energy 2011;36:7902e8.

[15] Nyamsi Serge Nyallang, Yang Fusheng, Zhang Zaoxiao. Anoptimization study on the finned tube heat exchanger usedin hydride hydrogen storage system - analytical method andnumerical simulation. Int J Hydrogen Energy2012;37:16078e92.

[16] Corgnale Claudio, Hardy Bruce J, Tamburello David A,Garrison Stephen L, Anton Donald L. Acceptability envelopefor metal hydride-based hydrogen storage systems. Int JHydrogen Energy 2012;37:2812e24.

[17] Nam Jinmoo, Ko Johan, Ju Hyunchul. Three-dimensionalmodeling and simulation of hydrogen absorption in metalhydride hydrogen storage vessels. Appl Energy2012;89:164e75.

[18] Dhaou H, Souahlia A, Mellouli S, Askri F, Jemni A, BenNasrallah S. Experimental study of a metal hydride vesselbased on a finned spiral heat exchanger. Int J HydrogenEnergy 2010;35:1674e80.

[19] Souahlia A, Dhaou H, Askri F, Mellouli S, Jemni A, BenNasrallah S. Experimental study and characterization ofmetal hydride containers. Int J Hydrogen Energy2011;36:4952e7.

[20] Botzung Maxime, Chaudourne Serge, Gillia Olivier,Perret Christian, Latroche Michel, Percheron-Guegan Annick,et al. Simulation and experimental validation of a hydrogenstorage tank with metal hydrides. Int J Hydrogen Energy2008;33:98e104.

[21] Yeong Ha Man. A numerical study of thermo-fluidphenomena in metal hydride beds in the hydridingprocessKim Kyu, Doo Song Ha, Sung Sikyong, Lee DongHyuk, editors. Int J Heat Mass Transf 2004;47:2901e12.

tal study of metal hydride-based hydrogen storage tank atrgy (2014), http://dx.doi.org/10.1016/j.ijhydene.2014.02.121