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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8

Available online at www.sciencedirect.com

ScienceDirect

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

Hydrogen charging in AX21 activated carbon-PCM-

metal foam-based industrial-scale reactor:
Numerical analysis

Atef Chibani a,b,*, Ghania Mecheri b, Slimane Merouani a, Aissa Dehane a

a
Laboratory of Environmental Process Engineering, Department of Chemical Engineering, Faculty of Process
Engineering, University Salah Boubnider Constantine 3, P.O. Box 72, 25000, Constantine, Algeria
b
Faculty of Sciences and Applied Sciences, Department of Process Engineering, Larbi Ben M'hidi University, 04000,
Oum El Bouaghi, Algeria

highlights

Numerical H2 storage in large-scale activated carbon (AC)-MF-PCM reactor was done.

Mass and heat transfer were analyzed in the various regions of the AC-MF-PCM reactor.

The MF-PCM system participates in the AC-charging through absorbing the reaction heat.

MFs considerably increases the melting rate of the pure paraffin (PCM).

Lower MF porosity accelerates the heat transfer and the PCM melting rate.

article info abstract

Article history: A computational fluid dynamics (CFD) model based on the mass, momentum, and energy
Received 12 November 2022 conservation equations is used to simulate the industrial-scale hydrogen storage in
Received in revised form activated carbon (AC: AX-21) bottle surrounded by a layer of metal foam (MF)-phase
20 February 2023 change material (PCM: RT 22 HC) latent thermal unit. The aim is to boost the recuper-
Accepted 6 March 2023 ation (storage) of the heat liberated from the H₂‘s exothermic (adsorption). Following the
Available online 18 May 2023 model's validation, several numerical simulations were performed to evaluate the
impact of various physical features of the MF (copper), at variable porosity (ε ¼ 0.80e1.0),
Keywords: on heat transfer in the hole reactor. Independently of the MF porosity, the AC-bed
Hydrogen storage temperature suddenly increased (from 295 K), reaching a maximum (at about 500 s)
Industrial-scale and decreased afterward. The temperature's intensity was dependent on the probe's
Adsorption position in the AC-bed (i.e. several positions were studied). The H2 concentration in the
PCM-MF unit AC-bed may reach 25 mmol/g. The PCM-based unit received the discharged heat from
Heat and mass transfer the AC-bed slowly, but with PCM incorporation in MF, the heat was quickly recovered
CFD (stored). Thanks to the MF-induced PCM's increased thermal conductivity, the heat flux,
the heat transfer coefficient, and the melting rate of the PCM were all accelerated (during
the charging phase) several times by decreasing the MF porosity (increasing MF mass)
from 1 (pure PCM) to 0.8. The temperature and the melting contours throughout the
absorption process (up 10,000 s) supported all these findings. It is obvious from the re-
sults of the current study that improving the reactor geometry and the solid bed's

* Corresponding author. Laboratory of Environmental Process Engineering, Department of Chemical Engineering, Faculty of Process
Engineering, University Salah Boubnider Constantine 3, P.O. Box 72, 25000, Constantine, Algeria.
E-mail address: chibaniatef@gmail.com (A. Chibani).
https://doi.org/10.1016/j.ijhydene.2023.03.049
0360-3199/© 2023 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
32026 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8
thermal conductivity (thermal performance) is necessary for an effective hydrogen
storage system.
© 2023 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Introduction
Nowadays, the supply of energy has become the backbone of
human life sustainability. To this end, the energy storage and
its development has been considered as an inexorable pro-
cess. The clean, sustainable, and excellent properties of
hydrogen make it a potential energy source for the future [1].
However, the storage of hydrogen is a hurdle that has to be
overcome in addition to its production and consumption [2].
Liquefaction, adsorption on adsorbent materials with huge
surface area, and absorption in metal hydrides are four crucial
methods for effective and secure hydrogen storage. Of course,
each strategy has its advantages and disadvantages. Govern-
ments have shown a remarkable interest towards the in-
vestment in hydrogen energies in order to broaden and
expand the energy storage and most importantly cushioning
the carbon emissions [3]. The analysis of the performance of
hydrogen cryo-adsorptive systems for automotive applica-
tions, reveals that a density of roughly 76 kg m3 for a 70 MPa
vessel and a driving distance of 150 km day1 are achievable.
In order to improve the industrial systems’ thermal efficiency,
the main concern of scientists is to reduce the entropy gen-
eration, entropy optimized flow of Darcy Forchheimer viscous
fluid with cubic autocatalysis chemical reactions was done by
Wang et al. [4].
Hydrogen storage in metal-hydrides (MH) is a secure and
efficient technique to store hydrogen (chemical storage). A lot
of research studies were recently done about this subject
(either experimentally or numerically) [5e18]. MHs are porous
materials of high level of safety, high H2 volumetric density,
ambient temperature and pressure functioning, and relatively
cheap-cost [12,19]. In parallel, the adsorption/desorption of
hydrogen onto other solid beds (e.g. Activated carbon) is
regarded as a promising means for an efficient physic storage
of hydrogen (easy reversible). According to Hermosilla Lara
et al. [20] (H2 adsorption in activated carbon), <22% is the
possible contribution of the adsorption in thermal generated
heat. As a result, the level of loaded hydrogen, the uptake
capacity, and adsorbent enthalpy all have a major role in
determining the peak temperature of the bed.
The heat produced by the adsorption process is less than
that produced by the compression effect, according to Xiao
et al. [21]. Moreover, for cylindrical reactors, axial direction heat
transmission is seen to be greater than radial route heat
transfer. Moreover, when the H2 mass flow is larger, the highest
temperature and balancing pressure are quickly attained. Ac-
cording to Momen et al. [22], the adsorption process is in charge
of approximately 24% of the temperature rise in the activated
carbon. If the charging cycle is carried out at freezing conditions
or with a better performed adsorbent, its participation may be
augmented. The adsorption/desorption cycles were examined
in our recent publication [23] in the presence of a PCM/oil-based
recovery/release system as a function of H2 flowrate (2.104 to
8.104 kg/s). It was discovered that when the mass flow rate of
hydrogen increased, the solid bed's capacity increased corre-
spondingly. Yet, the charging periods terminated at the same
time regardless of the H2 flowrate.
Phase change-materials (PCMs) have gained significant in-
terest in MH-based hydrogen storage reactors as heat
management-systems [24e32]. PCMs can replace the tradi-
tional HTF cooling/heating system, with the advantage of
acting dual role, i.e. heat absorbers during hydrogen charging
and heat suppliers during the discharging [24]. These materials
holding great latent heat (L, KJ/Kg) can stock important heat
and transfer it with solid-liquid phase change (melting/solidi-
fication) [33,34]. Numerous experimental and simulation
works have studied the heat transfer enhancement in PCMs
energy storage units by increasing their low thermal conduc-
tivity using highly conductive nano-oxides [35,36] and porous-
metal foams (MFs) [37e39]. This later (MFs) is one of the best
solutions adopted till now to accelerate the thermal perfor-
mance of latent thermal energy storage units. Ben Maad et al.
[40] quantitatively-investigated the use of LaNi5-MH with
LiNO3e3H2O-PCM to control the heat generated by hydrogen-
adsorption. 80.0% of the stored H2 in the MH can be dis-
charged using just the heat stored in the PCM during the
charging cycle. Using CNT nanoparticles effectively has been
shown to enhance thermal system performance for the goal of
improving building ventilation [4]. Mellouli's group [41e43] has
demonstrated that boosting the PCM conductivity by incor-
porating HTF tubes or incorporating metal foams may signifi-
cantly increase the performance of the chosen MH-PCM
storage configuration. Mghari et al. [31] reported that the MH-
PCM reactor favorably influenced the dehydriding/hydriding
period. According to their work, the latent-heat of the PCM had
a more significant effect on the system's reaction than the
thermal-conductivity. In contrast, Nyamsi et al. [44] revealed
that the PCM thermal conductivity effect outperforms that of
the latent heat. Ye et al. [24] suggested a new hydrogen storage
reactor with a concentric finned tube heat exchanger and PCM.
A numerical model is used to look at how things move and
react in two-reactors, with or without PCM. The analysis in-
dicates that the reactor encircled by PCM released heat and
absorbed hydrogen faster and took half as long to do so.
Alqahtani et al. [25] and Ye et al. [45] came up with a new
sandwich shape MH-PCM reactor design. The sandwich design
improved the rate of heat transfer, which drops the time
needed for absorption and desorption by greater than 80.0 and
70.0%, respectively, compared to the only PCM bed scheme
[25]. Other researchers have also confirmed the enhanced
thermal and storage performance of MH-PCM units [27,46,47].
Interestingly, investigations on efficient commercial bot-
tles using activated carbon as storage matrix are few despite
the great variety of reactor designs utilized in MH-PCM storage
systems. Most studies concentrate on small to medium-sized
 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8
tanks (either for H2-charging or-discharging). Literature is
particularly rare when the PCM unit is used with metal foams.
The design and modelling of large-scale AC-PCM-MF setups
are essential for comprehending the thermal behavior of
hydrogen storage and PCM response in such structures and for
extrapolating the lab-scale findings to the industry.
In this paper, the adsorption of hydrogen in activated
carbon is analyzed in the presence-of PCM embedded in
copper MF of different porosities. The entire (active carbon,
bed and MF-PCM) is used to study a large size bottle with
typical dimensions. Fluent software was used to developed a
computational bottle dynamics (CFD) model (based on finite
bottle-volume method) and study the behaviors of a AC-bed
and MF-PCM jacket.
Reactor configuration and mathematical model
The schematic diagram of the suggested activated carbon
hydrogen storage bottle is shown in Fig. 1. (right) illustrates
Fig. 1 e Schematic of physical model, computational domain and grid structure with MF-PCM-activated carbon based
reactor.
32027
the vertical cross-section of our hydrogen storage system. The
computational domain of the numerical simulation is given in
Fig. 1(left). The computational domain was symmetrical with
respect to the vertical axis. Therefore, only the right half of the
domain was simulated. The steel bottle height (H) is 150 cm
(1.50 m), with internal (R1) and external (R2) radii of 11.5 and
18.0 cm, respectively (Fig. 1(right). The feeding entrance of the
reactor is of 1.5 cm in radius and 10 cm in height (expansion
volume in Fig. 1(right). In this system, the AX-21 of activated
carbon is packed in the bottle's interne (of R1 ¼ 11.5 cm),
whereas the PCM (or MF-PCM with different porosity) systems
was loaded in the annular section (11.5 cm < R < 18 cm)
Fig. 1(right). It is worth mentioning that the used copper is
frequently adopted as MF because of their high thermal con-
ductivity [48e53]. For the temperature and heat flow moni-
toring, eight positions are selected along the axial axis (P1eP6)
and in the vicinity of the inner wall (from P8, P4, and P7) of the
reactor. Thermalephysical parameters of copper-MF in addi-
tion to those of the activated carbon, hydrogen, phase change
material (RT 22 HC), and the steel wall are shown in Table 1,
32028 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8

Mass conservation equation

Table 1 e Material properties [8,21,57,75e77].
In porous media, the mass conservation equation for
Activated Specific heat capacity (J/kg K) 825 hydrogen is given as [21,54]:
carbon Density (kg/m3) 517.6
Thermal conductivity (W/m K) 0.646


v εb rg
 vn
Porosity 0.49 þ V: rg !
v ¼  ð1  εb Þ rP MH2 (1)
vt vt
Hydrogen Density Ideal gas
Thermal conductivity (W/m K) 0.206 where, n (mol Kg1) is the absolute adsorbed amount per unit
Specific heat capacity (J/kg K) 14,700 activated carbon, εb the bed porosity of the activated carbon,
Viscosity (Pa s) 8.411e6 MH2 (Kg mol1) molecular mass of hydrogen, rg (Kg m3) the
Viscous resistance coefficient (m_2)
density of gaseous hydrogen, !
8.29055e7
v (m s1) the superficial velocity
Inertial resistance coefficient (m_1) 7586
Steel wall Density (kg/m3) 7830 v ¼ εb !
vector (! u , where !
u is the physical velocity) in porous
Specific heat capacity (J/kg K) 468 media, rp (Kg m3) the Particle density of activated carbon.
Thermal conductivity (W/m K) 13 On the right side of the aforementioned equation, the mass
RT 22 HC Density (kg/m3) solid liquid 760
source term simulates the mass moved from the gas phase to
(PCM) 700
the adsorbed phase per unit volume per second. This
Specific heat capacity (J/kg K) 2000
Thermal conductivity (W/m K) 0.20 parameter (the mass source term) is connected to the porosity
Viscosity (Pa s) 0.02 of the activated carbon bed and the adsorption rate.
Phase change temperature (K) 293.15e296.15
Latent heat of fusion (kJ/kg) 190 Momentum conservation equation
Metal foam Density (kg/m3) 8978 In the case of standard fluid flow, the momentum equation is
(copper) Specific heat capacity (J/kg K) 381
stated as [55]:
Thermal conductivity (W/m K) 387.6
v
!

r v þ V: rg !
v !
v ¼  Vp þ V:t þ rg !
g (2)
vt g
respectively. The loading of PCM in copper-MF is for acceler- !
In porous media case, a momentum source term S should be
ating heat transfer (heat recuperation) from the activated
placed on the right side of the above equation. The parameter
carbon (hot core due H2's adsorption) to the PCM matrix (the !
storage site). The adsorption velocity and temperature of t is the stress tensor and S (N m3) is the momentum source
hydrogen at the reactor entrance are fixed at 1 m/s and 295 K, term, which is based on the Ergun equation and includes
respectively. Similarly, at the beginning of the storage process, viscous and inertial losses. Its component in the “i” direction,
the MF-PCM body temperature is at the temperature of 295 K. “Si”, for simple homogeneous porous media, may be written as
The charging time was fixed at 500 s. The outside of our tank is [14]:
adiabatically insulated (Q ¼ 0). The hydrogen velocity at the  
m 1
bottle entrance is fixed at 1 m/s. Si ¼  vi þ C2 rg j!
v j vi (3)
k 2

The parameter m (Pa s) is the dynamic viscosity of hydrogen

gas, 1/k (m2) is the viscous resistance coefficient, C2 (m1) is
CFD model
the inertial resistance coefficient, j!
v j is the magnitude of the
For hydrogen storage velocity vector and vi is the velocity component along the i
direction.
The main foundation of the CFD model commonly resides in
mass., momentum and energy-conservation formulas of the Energy conservation equation
hydrogen system. Actually, the following formulas strongly The energy equation for the hydrogen storage tank represents
rely on hydrogen, activated carbon and a steel tank wall. It is the balance between the amount of energy collected in the
noteworthy to say that the adsorbed sub-models adopted in tank and the energy changes owing to convective flow, pres-
this work involves elements of Dubinin-Asatkhov (D-A) sure work, conductive and thermal dispersion fluxes, and heat
adsorption-isotherms and kinetics, which is applicable for up release due to adsorption process. It may be given as [56]:
to the critical point of hydrogen. In additions, the following v   

assumptions are made for resolving the transport phenome- εb rg Eg þ ð1  εb Þ rp Es þ V: ! v rg Eg þ p
vt !
non in the porous media. X ! (4)
¼ V: keff VT 
!
hi J i þ t: v þ Se
i
- The porous medium is assumed homogenous and
isotropic. The variable Se is an energy source parameter which is
- The solid-and gas temperatures are assumed to be equal expressed as [15]:
(thermal equilibrium-assumption). 
Se ¼  DHSm MH2 (5)
- Hydrogen is assumed to obey the perfect gas law (this
assumption is valid below 20 MPa). The parameter DH (KJ/Kmol) represents the constant
- The isosteric adsorption heat (DH) used in this study is isosteric heat of adsorption. The variation of DH is mainly due
assumed to be constant and equals to 10,500 J mol1. to the experimental circumstances as well as the structure of
 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8 32029

activated carbon. The energy source term (Se, W/m3) is related The phase change is accounted for by the porosity model
to the mass source term (Sm, kg m3 s1) and represents the that treats the transition zone as a porous zone with a liquid
heat's amount created per unit volume. Overall, for the mixing percentage of 0e1. In Eq. (2), the source term is defined as [59]:
features of the porous medium, the FLUENT™ software ex-
2
! 1  fl !
amines just two components. The specific heat is expressed as S ¼ A!
v ¼  Amush 3 v (12)
the volumetric average of the gas and solid phases: fl þ 0:001

 where Amush refers to the consecutive number in the mushy
rcp m
ps
¼ εb rg cpg þ ð1  εb Þrp c (6)
region. This constant is often between 105 and 107 [60]. The
approximation of Boussinesq is utilized, and Amush is taken as
ps ¼ cps þ qa cpa zcps þ qa cpg
c (7)
105 [60].
ps is the modified specific heat for the solid phase (adsor-
c Energy equations [61]:
bent and adsorbate), where the influence of the hydrogen  
vhsens ðTÞ dhlat ! !
adsorbed phase is included. r þ V:hsens ðTÞr!
v ¼ VðlVTÞ  r þ V :V:hlat (13)
vt dt

Adsorption models " ! #

v
 !

 ! ! X
To determine the link between hydrogen adsorption density rf Ef þ V: !
v rf Ef þ P ¼ V: leff V T  hi ji þ ðt:!
v Þ þ Se
and the pressure and the temperature in the tank, the modi- vt i

fied Dubinin-Astakhov_ (D-A) adsorption model is used. The (14)

absolute adsorption isotherm is given by Ref. [55]:
 m   p v2
RT p0 With : E ¼ h  þ (15)
n* ¼ nmax exp  ln
m
(8) r 2
a þ bT p
The total energy per unit mass (E) contained in the system
where, nmax (mol kg1) is the limit adsorption, the value of nmax
is comprised of three parts: internal, kinetic and potential.
is 71.6 mol kg1, m is a number generally close to a small integer
The sensible heat ℎsens(T), i.e. heat stored in the form of
and equal to 2 for most activated carbons. R is the universal gas
temperature increase in the liquid or the solid phase, is given
constant, and p is the equilibrium pressure. The value of the
as [62]:
saturation pressure p0 is 1470 MPa. The enthalpy a and the
entropic factors b are 3080 J mol1 and_18.9 J mol1 K1, ZT
respectively. The adsorption kinetics in the model is approxi- hsens ðTÞ ¼ href þ CP dT (16)
mated by the linear driving force (LDF) model [57]: Tm

dn href is the reference enthalpy at Tref ¼ 273 K and Cp is the

¼ k ðn*  nÞ (9) specific heat capacity at constant pressure. The energy stored
dt
during PCM phase transition, hlat(T), is given as [62]:
The parameter n* is the adsorbed gas in equilibrium with
the gas phase and k is a mass transfer coeffcient at an hlat ðTÞ ¼ f ðTÞ:DhSL (17)
aggregated level. In this paper, a value of k ¼ 0.45 s1 is
assumed. In addition, we select 10,500 J/mol as the constant where DHS-L is the total latent heat of the solid-liquid phase
isosteric heat of adsorption [57]. transition and ‘f' is the liquid fraction (0  f  1). f is given as
[52]:
CFD model for the MF-PCM thermal storage system 8
> 0 T  TS
>
>
<
T  Ts

For Pure PCM f¼ Ts  T  Tl (18)
>
> Tl  Ts
>
:
1 T  Tl
Voller's enthalpy-porosity formulation provided by the
software is used for simulating paraffin phase change. This subscripts “s" and “l" denote the PCM solid and liquid
model simplifies the phase change problem by providing the phases, respectively.
following benefits: (i) the governing equations are equivalent
to the equation for a unique phase, (ii) absence of condition to
For MF-PCM
meet the interface and (iii) the enthalpy formulation declares
a mixing zone between phases, permitting us to follow the For the MF-PCM system, the formulation of the constitutive
melting front easily. equations depends on the choice of the unknowns and the
Mass conservation [58]: assumptions made. By way of example, the three equations
adopted in the fluent thermophysical simulation code are as
!!
V: v ¼0 (10) follows:
Momentum conservation [58]: Mass conservation [63]

!
v! ! ! ! V: V ¼ 0 (19)
v Þrref !
v
rref þ ð!
v V:!
v Þ ¼ V P þVðmV:! g ð1 b:ðT Tm ÞÞ þ S
vt Momentum conservation [63]:
(11) In the x-direction
32030 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8

2 and continuity equations, respectively. For the entire melting

rpcm vu 1 ! vP mpcm
2  1  fl
þ V: V u ¼ þ V u  Amush 3 u process, 20 to 10 iterations are necessary for a convergent
ε vt ε vx ε fl þ 0:001
  solution in each cell and for each time step.
m rCpcm juj!
 þ pffiffiffiffi V u
K K
Time step and grid independency tests
(20)

In the y-direction To ensure the accuracy of simulation results, the computa-

2 tional domain's time step and grid resolution independence
rpcm vv 1 ! vP mpcm
2  1  fl are assessed. For the activated carbon-MF-PCM system (Fig. 3),
þ V: V v ¼ þ V v  Amush 3 v
ε vt ε vy ε fl þ 0:001 three time steps of 0.5, 0.75, 1 and 1.5 s four grid resolutions
 
m rCpcm jvj! (cell numbers) of 17,364, 9561, 7585 and 6388 cells with a fixed
 þ pffiffiffiffi V v þ ðrbÞpcm g:ðT  Tm Þ
K K time step of 1 s were selected to simulate the charging process
(21) of hydrogen. Figures (3a) and (3b) show the average temper-
Energy equations [63]: ature profiles of the activated carbon bed after three times
steps (with 9561 cells) and four times steps (with four cells).
h

 i vT
 !
ε rCp pcm þ ð1  εÞ rCp mf þ r rCp pcm V :V:T According to Fig. 3(a), the bed temperature rapidly increased
vt (22) during the initial loading phase (0e1000 s) due to the

 vf
¼ V leff VT  εðrLÞpcm : exothermal adsorption of H2 on the activated carbon, reaching
vt
a maximum (at 1000 s) before being quenched and gradually
In the thermal equilibrium model, volume average thermal returning to the initial temperature (after a longer time of
conductivity is assumed between the PCM and the porous 7000 s). An inappreciable deviation (5% max.) of the maximum
medium as [63]: bed temperature is shown between the four adopted numbers
of cells (17,364, 9561, 7585 and 6388), reflecting a strong sta-
leff ¼ ð1  εÞlMF þ εlpcm (23)
bility of the numerical solution. In parallel, the temperature
The initial and boundary conditions are defined as follows: profiles were stable for time steps of 0.75, 1, and 1.5 s, where a
At t ¼ 0, Ti ¼ 295.15 K, Vi ¼ 1 m/s, fpcm ¼ 0, n ¼ 0 mmol/g. maximum deviation of about 3e5% in the maximum bed
temperature was recorded (Fig. 3(b)). Therefore, throughout

Conduction/conduction heat continuity between AC and the remaining sections, a grid with a size of 9561 and a time
reactor wall: step of 1 s are used to shorten computing time and increase

 the accuracy of results.
leff VTAC  lw VTw :!
n ¼0 (24)
Model validation

Conduction/conduction heat continuity between PCM and
reactor PCM-MF It is quite effective to validate hydrogen storage models for

 various processing conditions and reactor designs using
lpcm VTpcm  lw VTw :!
n ¼0 (25) experimental data. Using experimental literature data from
Ye et al. [57], our constructed CFD model (Sections 3.1 and 3.2)


lpcm VTpcm  lMF VTMF :!
n ¼0 (26) is validated. With the identical experimental settings as Ye
et al. [57], i.e. reactor size, mass flow rates, Tambient ¼ 302 K,

The adiabatic (axis-symmetry) boundary condition: TH2,in ¼ 303 K, hwater ¼ 36 Wm-2 K1, Pinitial ¼ 0.04858 MPa, the
evolution of the bed temperature (activated carbon) is depic-
VTAC ¼ VTpcm ¼ 0 (27) ted during the charging, dormancy, and discharging time
(Fig. 2). The results of our model agree with those of Ye et al.
Numerical solution [57] by covering the adsorption, dormancy and the end of the
desorption phase (theoretical and practical outcomes). During
Energy charging is elaborately figured out using computa- the charging process, it is noticed that there is a heating. The
tional fluid dynamics (CFD) and finite volume discretization. temperatures are increased up to 335 K for the simulated (our
Explicitly, the equation's system (activated carbon-MF-PCM) results) and the experimental data (Fig. 2). The Ye et al. model
and their particular boundary conditions are solved under a marked maximum temperature for from of the experimental,
fixed grid computational domain. To successfully deal with 340 K. Up to 4000 s, our model accurately reproduces the
the pressure velocity coupling, a pressure based segregated observed temperature profiles; however, after this time, our
solver and a semi-implicit pressure linked equation are model underpredicts compared to Ye's team's model. The
adopted. For the sake of analyzing the pressure correction, the fundamental cause of this disagreement is the varied simu-
QUICK differencing method and PRESTO are established. lation circumstances (such as time step and geometrical dis-
Pressure, velocity and thermal energy all had under relaxation cretization, etc.) used in both researches. As a result, whether
factors of 0.3, 0.7 and 1, respectively. For time discretization, a PCM-based thermal process is present or not, our constructed
transitory formulation was adopted. 107, 105, and 105 were CFD model is reliable (correct) for predicting hydrogen
used as convergence conditions for the energy, momentum, charging in activated carbon.
 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8 32031

Fig. 2 e Validation results with the numerical and

experimental studies by the current model with those of
Ye et al. [57].

Results and discussion

Thermal analysis of hydrogen storage in the AC-MF-PCM

system

Fig. 4 (a)-(e) show the evolution of the temperature in points P1

to P8 in the activated carbon bed as a function of time in the
interval 0e10,000 s as the porosity of the thermal storage
material (RT22HC) varied from ε ¼ 100% (pure PCM, no MF) to
95, 90, 85 and 80% (i.e. the MF mass percentage increased from
0 to 5, 10, 15 and 20%, respectively). As it can be seen in Fig. 4,
during the adsorption phase (0e1000 s) of H2, the bed tem-
perature goes up, this, independently of the probing point
(P1eP8), whereas the intensity of temperature increase is
dependent on the position of each point (P1eP8). The increase
of bed temperature during the charging of H2 is ascribed to the
exothermicity of hydrogen adsorption (to the activated carbon
surface) as well as its compression process. It is worth
mentioning that according to Xiao et al. [21], the heat gener-
ated from the compression effect is greater than that from the
adsorption process, whereas the impact of adsorption is
intensified at the former period of H2 charging. Interestingly,
according to Fig. 4(a)e(e), inserting PCM into Cu-MF: (i) has no
effect of temperature profiles in point P1 to P4 (superimposed
profiles) and reduced the maximum pic temperature at points
P5, P6, P7 and P8 by ~ 5 K (P5: from 375 to 370 K, P6: from 330 to
about 325K, P7: from 350 to 345K and P8: from 420 to 415 K).
Thus, the bed temperature is position dependent. It is higher
for point 8 (at the bed center), followed by points 1e4 (same
vertical plan), then P5 followed by P6 and P7. The provided
profiles are in general porosity independent (little effect),
which means that changing the MF porosity has no appre-
ciable impact on the hydrogen adsorption matrix (the acti-
vated carbon bed). Similar statement has been reported early
for PCM loaded in MF coupled with H2 metal hydride storage
system [64]. This order of temperature distribution may be
Fig. 3 e Average bed (activated carbon) temperature for
different time steps (a) and grid size on independency tests
(b).
justified by the nearness or the farness of these points from
the bottle's center, as well as the strong impact of gas
compression inside the bottle. Of course, the temperature at
the central vertical of the bed could be maximal (reason for
why P8 achieved the elevated temperature profiles), but it also
depend on point height in the bottle (gravitational force's ef-
fect). In fact, during the adsorption period, the bottom region
of the reactor (from P1 to P4) is remarkably heated tanks to the
exothermic adsorption, compression effect (dominant-factor),
and the less evacuated amount of heat (to the PCM) in this
zone. P5 and P6 are located at upper part of reactor near the
entrance of the reactor. Of course, this zone could be filled
firstly by H2; the exothermic character could be more impor-
tant (highest-temperature!). Thus, for P5 and P6, more heat
energy is dissipated outside the AC bed. However, efficient
cooling simultaneously happens (at this zone) by the inlet H2
(295 K). Based on finding of Fig. 4 (a)e(e), the impact of the later
process overlapped that of the former, which provoked a low
temperature elevation at P5 and P6 [Fig. 4 (a)e(e)] as compared
32032 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8

Fig. 4 e Temperature monitoring points for different MF-porosities, ε ¼ 0.95, 0.90, 0.85, 0.80 and pure based metal foam-
PCM-AX-21 during adsorption process.

with P1eP4. Obviously, the lowest temperature is obtained at temperature until reaching the value ~300e320 K this
P7, this point is placed for from the center (just at the wall's decrease is mainly due to the sloppiness of the adsorption
AC-bed) which means that it easily loses its heat to the PCM process accomplished by the absorption of heat by the PCM as
system. I all case, beyond 1000 s, a sudden decrease in well as the inlet H2 induced cooling effect.
 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8
Hydrogen adsorption capacity and PCM melting degrees
The average quantity of hydrogen adsorbed (per unit mass of
AC) and the-melting percentage of PCM under the identical
circumstances as in Fig. 4 are presented in Fig. 5(a)-(b)
throughout the periods of 0e600 s and 0e10000s, respectively
(i.e. adequate time for both processes to stabilize). In terms of
hydrogen adsorption capability, changing the MF porosity
(from ¼ 100% [pure PCM] to 80%] has no noticeable effect. In
fact, a little effect doesn't really start to show up until 300 s in,
and after 600 s of operation, the adsorption capacity has only
increased by around 7% (24 mmol/g for 0,80e1 and 25 mmol/g
for ¼ (0,85e0,90)). This is because the heat absorption's rate by
the PCM is slow due to the very low PCM thermal conductivity.
These findings are in agreement with several studies from the
literature for various reactor designs. The high feeding ve-
locity (1 m/s), which causes a rapid sorption of the gas in
porous material, is the cause of the rapid thermal and kinet-
ical processes that take place in the AC bed [Fig. 5(a)e(b)]. As a
Fig. 5 e Adsorbed quantity and melt fraction for ε ¼ 0.95,
0.90, 0.85, 0.80 and pure based metal foam-PCM-AX-21
during adsorption process.
32033
result, the low area (mass) occupied by the PCM (30%) as
opposed to that of the AC metal (70%), might instead be used
to justify the MF porosity's independence from hydrogen
adsorption capability. The heat absorbed by the PCM,
regardless of its value, could not considerably affect the
thermal heat produced from the exothermic process of
hydrogen adsorption. This statement has been previously
stated for the case of hydrogen storage in metal hydride
storage material (chemisorption process).
The melt percentage of PCM, on the other hand, rose when
the MF's porosity decreased, as seen in Fig. 5(b). In the time
period of 0e1000 s, the melt fraction sharply rose to 12% (ε ¼ 1)
and 15e20% (ε s 1). The copper MF porosity clearly affects the
melt fraction after a prolonged period of operation. Melting
fractions of 24, 30, and 35% were observed at 4000 s while 30,
35, and 40% were reached at 10,000 s for ε a decrease from 1 to
0.9 and 0.85, successively. It is important to point out here the
slow recuperation rate of heat by the PCM (latent storage),
since the process continues to happen for 10,000 s. The MF-
PCM system improved heat transmission (and subsequently
melting rate), which is primarily attributable to the MF superior
PCM's thermal conductivity compared to PCM alone [37e39,64].
The MF-PCM system's conductivity might also be improved by
reducing the MF porosity [65,66], which would increase the
thermal exchange between the heated AC bed and the PCM
(heat storage system). However, because it reduces the PCM
amount, which is the latent heat's storage matrix, researchers
did not advise a drastic reduction the MF porosity [65,66].
MF's impact on heat flux and heat transfer coefficient
For the charging and cooling phases of the reactor, the influ-
ence of MF porosity on the heat transfer coefficient of the MF-
PCM system, the total surface heat flow, and the average bed
temperature (AC) are examined in Fig. 6(a)-(c) for 10,000 s.
According to Fig. 6(b), in the presence of MF and during the
loading process of H2, the heat transfer coefficient jumps to
~400 W m2 K1 (for ε ¼ (0,90e0,95)) and 460 W m2 K1 (for ε ¼
(0,80e0,85)), against ~55 Wm2K1 for pure PCM) at ~ 500 s,
then, it goes down rapidly at the end of the charging period
(1000 s). These data reflect an important boosting of the
thermal conductivity of the pure PCM with MF incorporation
at gradual decreased porosity (increased MF fraction from 5%
to 20%). Thereafter, the heat transfer coefficient fluctuated
until the end of the dormancy phase (10,000 s). The same trend
was observed for the surface heat flux with a lower-
magnitude, Fig. 6(a). The acquired trend in Fig. 6 (b) is a
direct translation of the AC-bed temperature evolution
(Fig. 6(c)). Due to H2 compression and adsorption, the AC-bed
temperature rose throughout the charging phase, resulting in
a significant heat flow (2800 W/m2 at 550s, Fig. 6(b)) being
evacuated towards the MF-PCM unit. In contrast to the
charging phase, the bed temperature progressively drops
during the resting interval (1000e10000 s). Our results are
consistent with many literature publications's results that
show that MF has a positive impact on the thermal perfor-
mance of the PCM in a variety of thermal unit's configurations
[37e39,64]. The results of Fig. 6 (b)e(c) explain well the melt
fraction behaviors of Fig. 5(b); the higher the heat flux received
by the MF-PCM block, the higher was the melting rate.
32034 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8
Fig. 6 e Average bed temperature, Heat transfer coefficient and surface heat flux of ε¼0.95, 0.90, 0.85, 0.80 and pure based
metal foam-PCM-AX-21 during adsorption process.
Consequently, massive stored heat (in the PCM material)
could be achieved. By considering the impact of heat transfer
coefficient on the heat flux and the bed temperature (AC),
Fig. 6(a)e(c), it can be noticed that according to the literature
findings, the variation of heat transfer coefficient (by the
change of HTF fluid) or the changing of the wall material have
negligible impact on the maximal temperature of adsorbent
[20,56,67]. This outcome was linked to the PCM solid bed's poor
heat conductivity. Consequently, future-research on H2
adsorptive-storage tanks should focus-primarily on
improving the solid bed's thermal conductivity.
Temperature and melt fraction contours of the MF-PCM
system
The 2D distribution of the temperature and PCM liquid frac-
tion during the paraffin melting in the AC-PCM-MF storage
unit can better illustrate the thermal storage issue and the
contribution of each reactor's part in the process. Fig. 7 show
these evolutions over the time interval ranging from 600 to
10,000 s. These-maps were determined in the absence (pure
PCM) or presence of MF at various porosities (ε ¼ 0.85e0.95).
This graphic shows that the same explanation holds true both
when MF is present and when it is not. The only exception is
that decreasing the porosity's-MF accelerated the displace-
ment of the liquid layer from the AC- wall to the MF-PCM
external wall. During the charging regime, maximum heat-
ing (T z 420 K at t ¼ 600 s) is observed at the bed center, where
the compression and H2 adsorption heats are hardly-
evacuated (outside the reactor) to the low thermal conduc-
tivity of solid bed and the-farness of its core from the inlet and
the enclosure's wall. In contrast, this temperature is gradually
decreased as we get closer to the reactor entrance where the
cooling effect of fresh H2 takes place. Due to the limited
thermal conductivity of the activated carbon, the bed core gets
substantially colder with a difficult dissipation of heat energy
as time passes (t ¼ 5000 and 10,000 s). Till the completion of
the dormancy phase (10,000 s), the temperature for the MF-
 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8 32035

PCM (or PCM alone) system remains nearly constant (295 K),
whereas a modest increase in temperature (299 K) is seen in
the area around the reactor wall. The liquid percentage of
PCM, where in each case the melt paraffin progressively grows
over time, serves as evidence that the temperature of MF-PCM
(or PCM) varies over time. Moreover, the thickness of liquid
PCM becomes somewhat wider at the top as time increases
(t > 600 s). This is clearly attributable to the reactor's top area
being heated greater during the dormant period. According to
the previous findings, the changes in bed temperature and
PCM liquid fractions demonstrate that the hydrogen storage
reactor's thermal performance and storage capacity are both
enhanced by using MFs [68e74].

Conclusion

The results of the present CFD study on hydrogen adsorption

on an activated carbon bed (AX-21 embedded in an industrial
bottle) surrounded by a PCM layer showed a significant
improvement in the PCM melting rate (enhanced heat recov-
ery) using copper metal foam. The adopted model was firstly
validated, and then specific numerical analysis was per-
formed to understand the thermodynamic behavior of the AC-
MF-PCM system at variable MF porosity. The main outcomes
are.

(i) During the adsorption phase (0e500s) of H2, the bed

temperature goes up independently of the probing point
(P1eP8), whereas the intensity of temperature increase
is dependent on the position of each point (P1eP8).
(ii) The heat generated by the compression effect is greater
than the heat generated by the adsorption process,
whereas the impact of adsorption is amplified during
the former period of H2.
(iii) The discharged heat from the exothermic adsorption of
hydrogen on GAC was efficiently stored as latent heat
into the PCM material.
(iv) The heat transfer coefficient was increased from 55 W/
m2 K for pure PCM to 400 W/m2 K (for ε ¼ 0,90e0,95) and
460 W/m2 K (for ε ¼ 0,80e0,85), respectively, during the
initial H2 charging cycle (around 500 s)

Finally, it is obvious from the results of the current study

that improving the reactor geometry and the solid bed's
thermal conductivity is necessary for an effective hydrogen
storage technology (on activated carbon). To achieve this,
further approaches, such as the incorporation of fins and/or
foams (of different shapes) into the solid bed body, could be
practicably implemented in addition to different reactor de-
signs. Additionally, involving Machine Learning or Artificial
Intelligence in the thermal storage field could be of great
importance for process optimization, control and prediction.

Declaration of competing interest

Fig. 7 e Liquid fraction and temperature contours, velocity
ε¼0.95, ε¼0.90, ε ¼0.85, ε¼0.80, pure based metal foam- The authors declare that they have no known competing
PCM-AX-21 at different time intervals while adsorption at financial interests or personal relationships that could have
constant inlet velocity of 1 m/s. appeared to influence the work reported in this paper.
32036 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8

references hydride hydrogen storage tanks for fuel cell vehicles. Int J
Hydrogen Energy 2010;35:1693e705. https://doi.org/10.1016/
j.ijhydene.2009.12.052.
[16] Mellouli S, Askri F, Dhaou H, Jemni A, Ben Nasrallah S.
[1] ezzahra Chakik F, Kaddami M, Mikou M. Effect of operating
Numerical study of heat exchanger effects on charge/
parameters on hydrogen production by electrolysis of water.
discharge times of metalehydrogen storage vessel. Int J
Int J Hydrogen Energy 2017;42:2e9. https://doi.org/10.1016/
Hydrogen Energy 2009;34:3005e17. https://doi.org/10.1016/
j.ijhydene.2017.07.015.
j.ijhydene.2008.12.099.
[2] Abdalla AM, Hossain S, Nisfindy OB, Azad AT, Dawood M,
[17] Tong L, Xiao J, Yang T, Pierre B. Complete and reduced
Azad AK. Hydrogen production, storage, transportation and
models for metal hydride reactor with coiled-tube heat
key challenges with applications: a review. Energy Convers
exchanger. Int J Hydrogen Energy 2019;44:15907e16. https://
Manag 2018;165:602e27. https://doi.org/10.1016/
doi.org/10.1016/j.ijhydene.2018.07.102.
j.enconman.2018.03.088.
[18] Oliva DG, Fuentes M, Borzone EM, Meyer GO, Aguirre PA.
[3] Zini G, Tartarini P. Wind-hydrogen energy stand-alone
Hydrogen storage on LaNi5 xSn x. Experimental and
system with carbon storage : modeling and simulation.
phenomenological Model-based analysis. Energy Convers
Renew Energy 2010;35:2461e7. https://doi.org/10.1016/
Manag 2018;173:113e22. https://doi.org/10.1016/
j.renene.2010.03.001.
j.enconman.2018.07.041.
[4] Wang F, Khan SA, Gouadria S, El-Zahar ER, Khan MI,
[19] Kim KJ, Lloyd G, Razani A, Feldman KT. Development of
Khan SU, Yasir M, Li Y-M. Entropy optimized flow of Darcy-
LaNi5/Cu/Sn metal hydride powder composites. Powder
Forchheimer viscous fluid with cubic autocatalysis chemical
Technol 1998;99:40e5. https://doi.org/10.1016/S0032-5910(98)
reactions. Int J Hydrogen Energy 2022;47:13911e20. https://
00088-6.
doi.org/10.1016/j.ijhydene.2022.02.141.
[20] Hermosilla-Lara G, Momen G, Marty PH, Le Neindre B,
[5] Busque  R, Torres R, Grau J, Roda V, Husar A. Mathematical
Hassouni K. Hydrogen storage by adsorption on activated
modeling, numerical simulation and experimental
carbon: investigation of the thermal effects during the
comparison of the desorption process in a metal hydride
charging process. Int J Hydrogen Energy 2007;32:1542e53.
hydrogen storage system. Int J Hydrogen Energy
https://doi.org/10.1016/j.ijhydene.2006.10.048.
2018;43:16929e40. https://doi.org/10.1016/ nard P, Chahine R. Simulation of
[21] Xiao J, Tong L, Deng C, Be
j.ijhydene.2017.12.172.
heat and mass transfer in activated carbon tank for
[6] Busque  R, Torres R, Grau J, Roda V. Effect of metal hydride
hydrogen storage. Int J Hydrogen Energy 2010;35:8106e16.
properties in hydrogen absorption through 2D-axisymmetric
https://doi.org/10.1016/j.ijhydene.2010.01.021.
modeling and experimental testing in storage canisters. Int J
[22] Momen G, Hermosilla G, Michau A, Pons M, Firdaous M,
Hydrogen Energy 2017;42:19114e25. https://doi.org/10.1016/
Marty P, Hassouni K. Experimental and numerical
j.ijhydene.2017.06.125.
investigation of the thermal effects during hydrogen
[7] Askri F, Ben Salah M, Jemni A, Ben Nasrallah S. A new
charging in packed bed storage tank. Int J Heat Mass Tran
algorithm for solving transient heat and mass transfer in
2009;52:1495e503. https://doi.org/10.1016/
metal e hydrogen reactor. Int J Hydrogen Energy
j.ijheatmasstransfer.2008.06.045.
2009;34:8315e21. https://doi.org/10.1016/
[23] Chibani A, Mecheri G, Dehane A, Merouani S, Ferhoune I.
j.ijhydene.2009.07.072.
Performance improvement of adsorptive hydrogen storage
[8] Chibani A, Bougriou C, Merouani S. Simulation of hydrogen
on activated carbon: effects of phase change material and
absorption/desorption on metal hydride LaNi5-H2: mass and
inconstant mass flow rate. J Energy Storage 2022;56:105930.
heat transfer. Appl Therm Eng 2018;142:110e7. https://
https://doi.org/10.1016/j.est.2022.105930.
doi.org/10.1016/j.applthermaleng.2018.06.078.
[24] Ye Y, Yue Y, Lu J, Ding J, Wang W, Yan J. Enhanced hydrogen
[9] Choi H, Mills AF. Heat and mass transfer in metal hydride
storage of a LaNi5 based reactor by using phase change
beds for heat pump applications. Int J Heat Mass Tran
materials. Renew Energy 2021;180:734e43. https://doi.org/
1990;33:1281e8.
10.1016/j.renene.2021.08.118.
[10] Cacciola G, Freni A, Cipitı F. Finite element-based
[25] Alqahtani T, Mellouli S, Bamasag A, Askri F, Phelan PE.
simulation of a metal hydride-based hydrogen storage tank.
Thermal performance analysis of a metal hydride reactor
2009. p. 8574e82. https://doi.org/10.1016/
encircled by a phase change material sandwich bed. Int J
j.ijhydene.2009.07.118. 34.
Hydrogen Energy 2020;45:23076e92. https://doi.org/10.1016/
[11] Nam J, Ko J, Ju H. Three-dimensional modeling and
j.ijhydene.2020.06.126.
simulation of hydrogen absorption in metal hydride
[26] Alqahtani T, Bamasag A, Mellouli S, Askri F, Phelan PE. Cyclic
hydrogen storage vessels. Appl Energy 2012;89:164e75.
behaviors of a novel design of a metal hydride reactor
https://doi.org/10.1016/j.apenergy.2011.06.015.
encircled by cascaded phase change materials. Int J
[12] Kyoung S, Ferekh S, Gwak G, Jo A, Ju H. Three-dimensional
Hydrogen Energy 2020;45:32285e97. https://doi.org/10.1016/
modeling and simulation of hydrogen desorption in metal
j.ijhydene.2020.08.280.
hydride hydrogen storage vessels. Int J Hydrogen Energy nard P, Yuan C, Xiao J. Hydrogen
[27] Tong L, Yuan Y, Yang T, Be
2015;40:14322e30. https://doi.org/10.1016/
release from a metal hydride tank with phase change
j.ijhydene.2015.03.114.
material jacket and coiled-tube heat exchanger. Int J
[13] Gambini M, Stilo T, Vellini M. High temperature metal
Hydrogen Energy 2021;46:32135e48. https://doi.org/10.1016/
hydrides for energy systems Part B : comparison between
j.ijhydene.2021.06.230.
high and low temperature metal hydride reservoirs. Int J
[28] Nguyen HQ, Shabani B. Thermal management of metal
Hydrogen Energy 2017;42:1e11. https://doi.org/10.1016/
hydride hydrogen storage using phase change materials for
j.ijhydene.2017.03.227.
standalone solar hydrogen systems: an energy/exergy
[14] Mohammadshahi SS, Gould T, Gray EMA, Webb CJ. An
investigation. Int J Hydrogen Energy 2022;47:1735e51. https://
improved model for metal-hydrogen storage tanks e Part 2:
doi.org/10.1016/j.ijhydene.2021.10.129.
model results. Int J Hydrogen Energy 2016;41:3919e27.
[29] Yao J, Zhu P, Guo L, Duan L, Zhang Z, Kurko S, Wu Z. A
https://doi.org/10.1016/j.ijhydene.2015.12.051.
continuous hydrogen absorption/desorption model for metal
[15] Mellouli S, Askri F, Dhaou H, Jemni A, Ben Nasrallah S.
hydride reactor coupled with PCM as heat management and
Numerical simulation of heat and mass transfer in metal
 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8
its application in the fuel cell power system. Int J Hydrogen
Energy 2020;45:28087e99. https://doi.org/10.1016/
j.ijhydene.2020.05.089.
[30] nard P, Chahine R. Thermal management of
Tong L, Xiao J, Be
metal hydride hydrogen storage reservoir using phase
change materials. Int J Hydrogen Energy 2019;44:21055e66.
https://doi.org/10.1016/j.ijhydene.2019.03.127.
[31] El Mghari H, Huot J, Xiao J, El H, Huot J, Xiao J. Analysis of
hydrogen storage performance of metal hydride reactor with
phase change materials. Int J Hydrogen Energy
2019;44:28893e908. https://doi.org/10.1016/
j.ijhydene.2019.09.090.
[32] El Mghari H, Huot J, Tong L, Xiao J. Selection of phase change
materials, metal foams and geometries for improving metal
hydride performance. Int J Hydrogen Energy
2020;45:14922e39. https://doi.org/10.1016/
j.ijhydene.2020.03.226.
[33] Zhou D, Zhao CY, Tian Y. Review on thermal energy storage
with phase change materials (PCMs) in building applications.
Appl Energy 2012;92:593e605. https://doi.org/10.1016/
j.apenergy.2011.08.025.
[34] Farid MM, Khudhair AM, Razack SAK, Al-Hallaj S. A review
on phase change energy storage: materials and applications.
Energy Convers Manag 2004;45:1597e615. https://doi.org/
10.1016/j.enconman.2003.09.015.
[35] Gorzin M, Hosseini MJ, Rahimi M, Bahrampoury R. Nano-
enhancement of phase change material in a shell and multi-
PCM-tube heat exchanger. J Energy Storage 2019;22:88e97.
https://doi.org/10.1016/j.est.2018.12.023.
[36] Kok B, Technical BK. Examining effects of special heat
transfer fins designed for the melting process of PCM and
Nano-PCM. Appl Therm Eng 2020;170:114989. https://doi.org/
10.1016/j.applthermaleng.2020.114989.
[37] Joshi V, Rathod MK. Thermal transport augmentation in
latent heat thermal energy storage system by partially filled
metal foam: a novel configuration. J Energy Storage
2019;22:270e82. https://doi.org/10.1016/j.est.2019.02.019.
[38] Xu Y, Ren Q, Zheng ZJ, He YL. Evaluation and optimization of
melting performance for a latent heat thermal energy
storage unit partially filled with porous media. Appl Energy
2017;193:84e95. https://doi.org/10.1016/
j.apenergy.2017.02.019.
[39] Liu Z, Yao Y, Wu H. Numerical modeling for solid-liquid
phase change phenomena in porous media: shell-and-tube
type latent heat thermal energy storage. Appl Energy
2013;112:1222e32. https://doi.org/10.1016/
j.apenergy.2013.02.022.
[40] Ben Ma ^ ad H, Miled A, Askri F, Ben Nasrallah S. Numerical
simulation of absorption-desorption cyclic processes for
metal-hydrogen reactor with heat recovery using phase-
change material. Appl Therm Eng 2016;96:267e76. https://
doi.org/10.1016/j.applthermaleng.2015.11.093.
[41] Mellouli S, Ben Khedher N, Askri F, Jemni A, Ben Nasrallah S.
Numerical analysis of metal hydride tank with phase change
material. Appl Therm Eng 2015;90:674e82. https://doi.org/
10.1016/j.applthermaleng.2015.07.022.
[42] Mellouli S, Abhilash E, Askri F, Ben Nasrallah S, Ben
Nasrallah S. Integration of thermal energy storage unit in a
metal hydride hydrogen storage tank. Appl Therm Eng
2016;102:1185e96. https://doi.org/10.1016/
j.applthermaleng.2016.03.116.
[43] Mellouli S, Askri F, Abhilash E, Ben Nasrallah S. Impact
of using a heat transfer fluid pipe in a metal hydride-
phase change material tank. Appl Therm Eng
2017;113:554e65. https://doi.org/10.1016/
j.applthermaleng.2016.11.065.
[44] Nyamsi SN, Tolj I, Lototskyy M. Metal hydride beds-phase
change materials: dual mode thermal energy storage for
32037
medium-high temperature industrialwaste heat recovery.
Energies 2019;12. https://doi.org/10.3390/en12203949.
[45] Ye Y, Lu J, Ding J, Wang W, Yan J. Numerical simulation on
the storage performance of a phase change materials based
metal hydride hydrogen storage tank. Appl Energy
2020;278:115682. https://doi.org/10.1016/
j.apenergy.2020.115682.
[46] Rabienataj Darzi AA, Hassanzadeh Afrouzi H, Moshfegh A,
Farhadi M. Absorption and desorption of hydrogen in long
metal hydride tank equipped with phase change material
jacket. Int J Hydrogen Energy 2016;41:9595e610. https://
doi.org/10.1016/j.ijhydene.2016.04.051.
[47] Garrier S, Delhomme B, de Rango P, Marty P, Fruchart D,
Miraglia S. A new MgH2 tank concept using a phase-change
material to store the heat of reaction. Int J Hydrogen Energy
2013;38:9766e71. https://doi.org/10.1016/
j.ijhydene.2013.05.026.
[48] Sushobhan BR, Kar SP. Thermal modeling of melting of nano
based phase change material for improvement of thermal
energy storage. Energy Proc 2017;109:385e92. https://doi.org/
10.1016/j.egypro.2017.03.035.
[49] Eanest Jebasingh B, Valan Arasu A. A detailed review on heat
transfer rate, supercooling, thermal stability and reliability of
nanoparticle dispersed organic phase change material for
low-temperature applications. Mater Today Energy
2020;16:100408. https://doi.org/10.1016/j.mtener.2020.100408.
[50] Barthwal M, Dhar A, Powar S. Effect of nanomaterial
inclusion in phase change materials for improving the
thermal performance of heat storage: a review. ACS Appl
Energy Mater 2021;4:7462e80. https://doi.org/10.1021/
acsaem.1c01268.
[51] Chibani A, Merouani S. Acceleration of heat transfer and
melting rate of a phase change material by nanoparticles
addition at low Concentrations. Int J Thermophys
2021;42:1e16. https://doi.org/10.1007/s10765-021-02822-z.
[52] Khan Z, Khan ZA, Sewell P. Heat transfer evaluation of metal
oxides based nano-PCMs for latent heat storage system
application. Int J Heat Mass Tran 2019;144:118619. https://
doi.org/10.1016/j.ijheatmasstransfer.2019.118619.
[53] Khodadadi JM, Fan L, Babaei H. Thermal conductivity
enhancement of nanostructure-based colloidal suspensions
utilized as phase change materials for thermal energy
storage: a review. Renew Sustain Energy Rev 2013;24:418e44.
https://doi.org/10.1016/j.rser.2013.03.031.
[54] nard P, Chahine R. Charge-discharge cycle
Xiao J, Be
thermodynamics for compression hydrogen storage system.
Int J Hydrogen Energy 2016;41:5531e9. https://doi.org/
10.1016/j.ijhydene.2015.12.136.
[55] Xiao J, Peng R, Cossement D, Be nard P, Chahine R. Heat and
mass transfer and fluid flow in cryo-adsorptive hydrogen
storage system. Int J Hydrogen Energy 2013;38:10871e9.
https://doi.org/10.1016/j.ijhydene.2013.04.042.
[56] Xiao J, Tong L, Cossement D, Be nard P, Chahine R. CFD
simulation for charge-discharge cycle of cryo-adsorptive
hydrogen storage on activated carbon. Int J Hydrogen Energy
2012;37:12893e904. https://doi.org/10.1016/
j.ijhydene.2012.05.079.
[57] Ye F, Xiao J, Hu B, Benard P, Chahine R. Implementation for
model of adsoptive hydrogen storage using UDF in fluent.
Phys Procedia 2012;24:793e800. https://doi.org/10.1016/
j.phpro.2012.02.118.
[58] Mahdi JM, Nsofor EC. Solidification of a PCM with
nanoparticles in triplex-tube thermal energy storage system.
Appl Therm Eng 2016;108:596e604. https://doi.org/10.1016/
j.applthermaleng.2016.07.130.
[59] Chibani A, Merouani S, Bougriou C, Hamadi L. Heat and mass
transfer during the storage of hydrogen in LaNi5-based metal
hydride: 2D simulation results for a large scale, multi-pipes
32038 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 e n e r g y 4 8 ( 2 0 2 3 ) 3 2 0 2 5 e3 2 0 3 8
fixed-bed reactor. Int J Heat Mass Tran 2020;147:118939.
https://doi.org/10.1016/j.ijheatmasstransfer.2019.118939.
[60] Chibani A, Merouani S, Benmoussa F, Abdellattif MH, Erto A,
Jeon BH, Benguerba Y. A strategy for enhancing heat transfer
in phase change material-based latent thermal energy
storage unit via nano-oxides addition: a study applied to a
shell-and-tube heat exchanger. J Environ Chem Eng
2021;9:106744. https://doi.org/10.1016/j.jece.2021.106744.
[61] Chibani A, Merouani S, Gherraf N, Ferhoune I, Benguerba Y.
Numerical investigation of heat and mass transfer during
hydrogen desorption in a large-scale metal hydride reactor
coupled to a phase change material with nano-oxide
additives. Int J Hydrogen Energy 2022;47:14611e27. https://
doi.org/10.1016/j.ijhydene.2022.02.171.
[62] Kok B. Examining effects of special heat transfer fins
designed for the melting process of PCM and Nano-PCM.
Appl Therm Eng 2020;170:114989. https://doi.org/10.1016/
j.applthermaleng.2020.114989. Contents.
[63] Chibani A, Merouani S, Bougriou C. The performance of
hydrogen desorption from a metal hydride with heat supply
by a phase change material incorporated in porous media
(metal foam): heat and mass transfer assessment. J Energy
Storage 2022;51:104449. https://doi.org/10.1016/
j.est.2022.104449.
[64] Chibani A, Merouani S, Benmoussa F. Computational
analysis of the melting process of Phase change material-
metal foam-based latent thermal energy storage unit: the
heat exchanger configuration. J Energy Storage
2021;42:103071. https://doi.org/10.1016/j.est.2021.103071.
[65] Sundarram SS, Li W. The effect of pore size and porosity on
thermal management performance of phase change material
infiltrated microcellular metal foams. Appl Therm Eng
2014;64:147e54. https://doi.org/10.1016/
j.applthermaleng.2013.11.072.
[66] Mesalhy O, Lafdi K, Elgafy A, Bowman K. Numerical study for
enhancing the thermal conductivity of phase change
material (PCM) storage using high thermal conductivity
porous matrix. Energy Convers Manag 2005;46:847e67.
https://doi.org/10.1016/j.enconman.2004.06.010.
[67] Delahaye A, Aoufi A, Gicquel A, Pentchev I. Improvement of
hydrogen storage by adsorption using 2-D modeling of heat
effects. AIChE J 2002;48:2061e73. https://doi.org/10.1002/
aic.690480919.
[68] Chen J, Jiang J, Ma A, Song D. Research progress of phase
change materials (PCMs) embedded with metal foam
(a review). Procedia Mater. Sci. 2014;4:369e74. https://doi.org/
10.1016/j.mspro.2014.07.579.
[69] Sardari PT, Babaei-Mahani R, Giddings D, Yasseri S,
Moghimi MA, Bahai H. Energy recovery from domestic
radiators using a compact composite metal Foam/PCM latent
heat storage. J Clean Prod 2020;257:120504. https://doi.org/
10.1016/j.jclepro.2020.120504.
[70] Buonomo B, Manca O, Nardini S, Plomitallo RE. Numerical
study on latent heat thermal energy storage system with
PCM partially filled with aluminum foam in local thermal
equilibrium. Renew Energy 2022;195:1368e80. https://
doi.org/10.1016/j.renene.2022.06.122.
[71] Buonomo B, Celik H, Ercole D, Manca O, Mobedi M.
Numerical study on latent thermal energy storage systems
with aluminum foam in local thermal equilibrium. Appl
Therm Eng 2019;159:113980. https://doi.org/10.1016/
j.applthermaleng.2019.113980.
[72] Esapour M, Hamzehnezhad A, Rabienataj Darzi AA,
Jourabian M. Melting and solidification of PCM embedded in
porous metal foam in horizontal multi-tube heat storage
system. Energy Convers Manag 2018;171:398e410. https://
doi.org/10.1016/j.enconman.2018.05.086.
[73] Chen Z, Gao D, Shi J. Experimental and numerical study on
melting of phase change materials in metal foams at pore
scale. Int J Heat Mass Tran 2014;72:646e55. https://doi.org/
10.1016/j.ijheatmasstransfer.2014.01.003.
[74] Mahdi JM, Mohammed HI, Hashim ET, Talebizadehsardari P,
Nsofor EC. Solidification enhancement with multiple PCMs,
cascaded metal foam and nanoparticles in the shell-and-
tube energy storage system. Appl Energy 2020;257:113993.
https://doi.org/10.1016/j.apenergy.2019.113993.
[75] Chibani A, Bougriou C. Effect of the tank geometry on the
storage and destocking of hydrogen on metal hydride
(LaNi5H2). Int J Hydrogen Energy 2017;42:23035e44. https://
doi.org/10.1016/j.ijhydene.2017.07.102.
[76] Afzal M, Gupta N, Mallik A, Vishnulal KS, Sharma P.
Experimental analysis of a metal hydride hydrogen storage
system with hexagonal honeycomb-based heat transfer
enhancements-part B. Int J Hydrogen Energy
2021;46:13131e41. https://doi.org/10.1016/
j.ijhydene.2020.11.275.
[77] Xiao J, Hu M, Benard P, Chahine R. Simulation of hydrogen
storage tank packed with metal-organic framework. Int J
Hydrogen Energy 2013:13000e10. https://doi.org/10.1016/
j.ijhydene.2013.03.140.