Journal of Energy Storage 84 (2024) 110803

Contents lists available at ScienceDirect

Journal of Energy Storage

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

Research papers

Numerical investigation and response surface optimization of a sorption

heat storage systems performance using Y-shaped fins
Elham Abohamzeh *, Seyed Ehsan Hosseinizadeh , Georg Frey
Automation and Energy Systems, Saarland University, 66123 Saarbrücken, Germany

A R T I C L E I N F O A B S T R A C T

Keywords: Several previously developed prototypes have proven the high energy density of sorption heat storage systems.
Adsorption However, the very poor thermal conductivity of the material bed, especially in closed sorption storage systems,
Y-shape fins has remained one of the main challenges for introducing this technology to the market. In this paper, the use of
Sorption heat storage
straight and Y-shaped fins is investigated to break through the limitation of heat transfer. An axisymmetric
Numerical simulation
transient analysis is established for the adsorption and desorption processes of zeolite/water pair in an
Optimization
Response surface method adsorption bed containing Y-shaped fins. The results of numerical simulations reveal higher adsorption rates, and
improved heat transfer throughout the bed and also between zeolite and heat transfer fluid using Y-shaped fins
during the adsorption process, leading to increasing outlet temperature of the heat transfer fluid. The effect of Y-
shaped fin length and the angle between the branches has been investigated through a parametric study, aiming
to find the optimized fin geometry. The response surface method (RSM) has been applied to relate the achieved
outlet temperature to the fin geometry. The investigations were carried out independent of the amount of metal
used, to study the improvement only as a result of changing the geometry and arrangement of fins. Based on the
investigations, the optimized geometry of the Y-shaped fins has been defined. The average temperature lift for a
time period of 800 min has increased by 4.6 K compared to the case with straight fins and 8.7 K compared to the
case without fins, and for the average output power, 103 % and 238.8 % improvement is observed compared
with the case with straight fins and without fins, respectively. The result of the study for the desorption process
shows an increase of 42 K in the temperature of the adsorbent bed for the optimum case compared with the case
without fins.

considering its variability with location, time, and seasonal availability.

Using this storage system, the free solar energy can be stored for periods
1. Introduction
of low solar irradiation. These systems are considered a promising
technology to minimize the temporal mismatch between solar energy
The building sector constitutes 40 % of global energy consumption. It
supply and heat demand. Compared with other heating storage ap­
also contributes 30 % of greenhouse gases yearly [1]. Furthermore, the
proaches, latent and sensible, the noticeable advantage of this technol­
energy needed for cooling and heating is the highest energy consump­
ogy is its long-term energy conservation with high energy density and
tion in buildings [2]. With a 1.4 % increase in global energy-related CO2
negligible heat loss. These properties give the process a very advanta­
emissions, a historic high of 32.5 Gt was observed in 2017 [3].
geous and attractive long-term energy storage at both high and low
Considering this, novel technologies like adsorption heating and cooling
temperatures [4]. On the other hand, in adsorption cooling systems, the
systems for buildings have attracted a lot of attention, recently. The
cooling effect is desired and obtained through the reactions between two
process in these systems is characterized by a reversible reaction of re­
substances.
actants and products that involves an exothermic synthesis and endo­
The performance of both cooling and heating sorption systems is
thermic decomposition process. The adsorbed and released heat from
largely related to the mass and heat transfer inside the adsorbent bed.
these reactions can be considered for cooling and heating purposes.
Considering this, the improvement of these processes has been the
For heating applications, sorption storage systems have been pro­
subject of several recently conducted studies [5–7]. The focus of these is
posed as a solution for storing solar heat in summer and using the stored
mostly the materials characterization and reaction kinetics, and the
energy in winter, given the fact that using solar energy is limited

* Corresponding author.
E-mail address: elham.abohamzeh@aut.uni-saarland.de (E. Abohamzeh).

https://doi.org/10.1016/j.est.2024.110803
Received 17 April 2023; Received in revised form 30 January 2024; Accepted 31 January 2024
Available online 15 February 2024
2352-152X/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Nomenclature Greek symbols

ρ Density,kg/m3
C Water vapor concentration, mol/m3 ε Porosity
Cp Specific heat capacity, J/kg⋅K κ Permeability, m2
Ea Surface diffusion activation energy, J/mol μ Viscosity, kg/m⋅s
h Heat transfer coefficient, W/m2 ⋅K
hti Heat transfer coefficient between heat transfer fluid and Subscripts
bed Adsorption bed
tube, W/m2 ⋅K
co Copper
hto Heat transfer coefficient between heat adsorbent and tube,
con Condenser
W/m2 ⋅K
Evap Evaporator
ΔH Enthalpy of reaction, kJ/kg
eff Effective
k Thermal conductivity,W/m⋅K
EQ Equilibrium
Mw Average molar mass of water, kg/mol
f Water vapor
P Pressure, pa
htf Heat transfer fluid
R Universal ideal gas constant, J/mol⋅K
opt Optimized
T Temperature,K
p Particle
u Water vapor velocity vector, m/s
s Adsorbent, zeolite
X Amount of Adsorbed water vapor per unit mass of the
sf Straight fin
adsorbent, kgw /kgs
sat Saturation
XEQ Equilibrium loading, kgw /kgs
w Without fin
V Volume, m3

improvement of thermal conductivity of sorption materials. A 2D model completion of the adsorption process, this trend was reversed, with the
of an adsorption cooling system has been developed by Çağlar [8] to center exhibiting the highest and the outer surface showing the lowest
compare the performance with the finned and finless tube-type adsor­ rate. Crespo et al. [15] conducted a study aimed to optimize the dis­
bent bed using a water/silica gel pair. A considerable improvement in charging process of a thermal energy storage system for residential use.
the adsorbent bed temperature has been observed using the finned tube- The researchers discovered that employing conservative and consistent
type adsorbent bed. Mitra et al. [9] studied the effect of the aspect ratio discharging temperature set points resulted in a 9 % increase in the
of the heat exchanger as well as the size of adsorbent particles size on the energy density, as it experienced fewer interruptions during discharge.
characteristics of the adsorption process. They have considered two Additionally, the study explored the influence of weather conditions and
different particle sizes and evaluated three domains of heat exchanges. various low-temperature heat sources. Overall, the study concluded that
The domains had various aspect ratios but the same area. Based on the an energy density of 139 kWh/m3 can be achieved with a storage system
results from their CFD study, the dynamic uptake is strongly dependent comprising 20 modules of LiCl-silica gel when a constant low-
on the porous media flow resistance for smaller particle sizes, while a temperature heat source, such as waste energy or geothermal, was
weak dependence was observed on thermal and intraparticle diffusion available.
for particles with larger sizes. Some other studies have focused on Mohammad et al. [16] performed a numerical and experimental
optimizing the adsorbent bed for sorption systems. Kant et al. [10] study on an adsorption cooling systems performance and proposed a
developed a numerical model to investigate the silica gel/water modular packed bed. Their design showed 2.3 times greater specific
adsorption process in the adsorbent bed containing cylindrical as well as cooling power compared to typical adsorption chillers. Considering the
rectangular fins for sorption heat storage purposes. They also conducted advantages of this technology, research in the field of sorption storage
a parametric study on fin parameters. The results of their work showed has experienced rapid development, in recent years [17]. There is an
better performance of the rectangular fin compared with the cylindrical increasing number of research teams in engineering, chemical, and
one. Golparvar et al. [11] studied the influence of spacing of the fins in a materials sciences who concentrate on developing this technology [18].
gas-driven adsorption cooling system. Based on their results, higher Regarding the materials used for the sorption process, several studies
temperature gradients have been observed for a bed with larger fin have been conducted in recent years [19,20]. The two most commonly
spacings, leading to non-uniform desorption and adsorption processes, used traditional physical sorbents are silica gel and zeolite. Silica gel is a
so decreasing the performance of the adsorption cooling system. Kant widely and commercially used, cheap and non-toxic. However, the
and Pitchumani [12] conducted a numerical investigation to investigate storage capacity in practice is significantly lower than was theoretically
the influence of fin tree structure on improving heat transfer in a ther­ predicted and the sorption heat is also too low, as shown by Jaehnig
mochemical heat storage bed. They optimized the geometry of the fin for et al. [21]. Zeolites consist of aluminum silicate minerals, which have
maximizing energy storage density and minimize the levelized cost. very small porous structures. The pores allow them to absorb water and
Reichl et al. [13] investigated two approaches for numerically charac­ store it temporarily. In contrast to silica gel, they have a high desorption
terizing adsorption reactors in open sorption systems. The mixing temperature between 150 ◦ C and 300 ◦ C. Furthermore, zeolite has a
behavior of a rotating drum configuration is assessed using discrete higher energy density of 3300–4200 kJ/kg [22]. There are a large
particle models and particle simulations. The CFD simulations revealed number of different types of zeolites. Four species are particularly
enhanced reactor performance by improving particle mixing and prevalent in sorption storage. These are Zeolites 4A, 5A, 10X and 13X
reducing temperature differentials. Kurniawan and Rachmat [14] [23]. Herzog et al. [24] studied binderless molecular sieves and iden­
developed a new design for an adsorption chiller with silica gel/water tified A, X, and Y molecular sieves as potential low-temperature storage
pair, incorporating two sorption chambers with compact fin tube heat materials (up to 200 ◦ C). Zeolite A is particularly suitable for higher
exchangers, and conducted a CFD simulation. Based on the results, the temperatures because it has high thermal stability. For applications
outer surface of the adsorbent bed exhibited the highest absolute under high pressure in a reactor, for example, when working with
adsorption rate, while the center displayed the lowest rate. Upon condensing water, the Y variants are more suitable than the A and X

2
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 1. The schematic showing (a) the adsorbent bed and (b) the geometry of the Y-shaped fin.

Fig. 2. A schematic of the computational domain

variants. Type X, on the other hand, has better kinetic properties. Type A techniques are collected in RSM for optimization purposes. RSM was
produces the most usable temperatures. initially developed for modeling experimental results and after that used
In this work, the response surface methodology (RSM) is used for to model numerical experiments. Inaccuracy in physical experiments
optimizing the fin geometry. A series of statistical and mathematical can occur from errors in measurements, but numerical inaccuracy can be

3
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Table 1 ratio improves the thermal-hydraulic performance, whereas the oppo­

Geometrical parameters of the system site impact is seen at higher water volumetric flow rate or lower air
Parameter Value Unit Description velocity. Alshihmani et al. [29] conducted experiments to investigate
the impact of phase change material (PCM) on a printed circuit board's
Lbed 300 mm Length of the bed
ro 58 mm Outer Radii of the bed (PCB) performance. The cooling effects of two different PCM types,
ri 7 mm Inner Radii of the bed RT42 and RT35/CH, with various melting points and latent heat of
t 1 mm Tube thickness fusion, are compared under both free and forced convection. The study
L 10–20 mm Length of fin conducts experiments at two heat fluxes and utilizes the response sur­
b 4.45–8.6 mm Thickness of fin
α 0–175 − Angle between branches
face method (RSM) to develop predictive models for cooling perfor­
mance. The results show that using PCM improves module usage time
and enhancement ratio, particularly at low heat flux. Increasing PCM
volume fraction enhances usage time, with RT42 performing better
Table 2 overall. RSM highlights the significant impact of PCM volume fraction
Thermophysical properties of the materials on cooling performance compared to fan voltage.
Parameter Value Unit Description Although the systems working based on adsorption and desorption
dp 2 mm Particle diameter [31]
processes seem to be a very attractive option for air conditioning ap­
εbed 0.37 – Porosity of desorption bed [32,33] plications in buildings considering their high energy density, the poor
εp 0.42 – Porosity of zeolite Particle [32,33] thermal conductivity in the bed limits the performance of these systems.
n 1.73 – Heterogeneity factor The focus of this study is developing a coupled mass and heat mathe­
ΔH 3300 kJ/kg Iso-steric heat of adsorption [31] matical model for the simulation of the ad- and desorption process inside
R 8.314 J/(mol⋅K) Ideal gas constant
Mw 18.02 g/mol Water vapor molar mass
the zeolite bed and improving the process using Y-shaped fins. A para­
Ea 10,000 J/mol Surface diffusion activation energy - zeolite metric study has been performed to investigate the influence of geo­
kc kco 400 W/(m⋅K) Thermal conductivity of copper metric parameters of the fin on the adsorption process. The aim is to
ks 0.1 W/(m⋅K) Thermal conductivity of zeolite define the most efficient design of fins that can overcome the poor
8700 kg/m3 Density of copper
thermal conductivity of the zeolite bed. The optimized design has been
ρc co
ρs 1000 kg/m3 Zeolite density
Cp,c Cp,co 385 J/(kg⋅K) Specific heat of copper proposed using the RSM analysis. The system performance during both
Cp,s 837 J/(kg⋅K) Specific heat of zeolite the desorption and adsorption processes has been investigated.
Cp,htf 1930 J/(kg⋅K) HTF specific heat
khtf 0.115 W/(m⋅K) HTF thermal conductivity 2. Physical model
D0 5.8e− 9 m2/s Reference diffusivity
ρp,htf 914 kg/m3 HTF density
In this work, a cylindrical fixed bed reactor is considered as the
research object in a closed sorption heat storage system, as depicted in
Fig. 1. The working principle of the system during the adsorption pro­
Table 3 cess can be explained as follows: the process begins with the evaporation
Considered values for initial and boundary conditions of water. The generated water vapor flows into the bed and is adsorbed
Parameter Value Unit Description in the pores of dry zeolite so that the adsorption energy is released. The
Pin 3 Pa Initial pressure released heat is extracted by the heat transfer fluid (HTF) flowing
Tevap 278.15–293.15 K Evaporator temperature through the heat exchanger and is utilized for space heating and do­
Tinit 303.15 K Initial temperature mestic hot water. During desorption, the saturated zeolite is dried by the
heat provided by hot HTF flowing into the heat exchanger tube.
Consequently, the water is released in the gas form, flowing into the
due to the discrete representing of a continuous physical phenomenon,
condenser and condensing to the liquid [4,30].
round-off error, or incomplete convergence of iterative processes. The
The computational domain including the bed filled with zeolite 13X,
use of RSM in the optimization of designs aims to lower the expense of
the tube containing HTF and fins is presented in Fig. 2. The inner and
costly analysis techniques (such as CFD analysis or finite element
outer tube radii, ri and ro, the length of the heat exchanger, Lbed, and the
method) and the associated numerical noises [25]. Liu and Yang [26]
thickness, t, are defined as 7 mm, 58 mm, 300 mm, and 1 mm, respec­
used RSM to define the impact of different parameters on the perfor­
tively. Y-shaped fins made of copper are evenly distributed on the sur­
mance of concrete thermal energy storage. The results of the study
face. The thermophysical properties of water vapor, fin, zeolite,
identified the velocity of the heat transfer fluid as the most influential
geometrical, and other parameters are presented in Tables 1, 2, and 3
factor for energy efficiency and charging time. Furthermore, the inlet
[31,32].
temperature of heat transfer fluid significantly affected energy storage.
For the length of the fins a range of 10–20 mm and for the angle
An optimization process using the desirability function was performed
between the branches a range of 0–175◦ has been considered, to provide
to maximize energy storage efficiency and minimize charging time.
the opportunity for the investigation in a wide range. The thickness of
Ahmed et al. [27] designed and tested a lab-scale prototype of an
the fins is calculated based on the selected length so that the volume of
innovative concentrated solar-powered flash desalination system. They
the fins remains constant for both straight and Y-shaped fins. Hence, the
used RSM for the system modeling and optimizing the developed system
results are independent of the amount of used metal. It should be also
and investigated the influence of vacuum pressure, feed water temper­
mentioned, that this point is also considered for the straight fin type; the
ature, and flow rate on distillate production. Sun and Zhang [28] pro­
thickness and the length of the straight fin are considered so that the
posed a numerical model for predicting the heat transfer and fluid flow
volume of the fin would be the same as Y-shaped fins.
on both the water-side and air-side of the elliptical finned-tube heat
exchanger. RSM was employed to investigate the influence of seven
3. Mathematical model
considered design factors, including the water volumetric flow rate, air
velocity, fin pitch, longitudinal and transversal tube pitch, axis ratio,
3.1. Adsorption model for zeolite 13X
and row number. The findings show that water volumetric flow rate and
air velocity strongly interact with the axis ratio. When the water volu­
The linear driving force (LDF) model is utilized for modeling the
metric flow rate is lower or air velocity is higher, increasing the axis
adsorption rate and solving the adsorbed amount [34].

4
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 3. Grid independence test.

∂X This theory says that the adsorbed amount is governed by the micropore
= K LDF (XEQ − X) (1)
∂t volume rather than the surface area of the micropores. The equilibrium
water uptake is the theoretical maximum value of the adsorbate which
where XEQ is the adsorption capacity in the equilibrium condition and X could be adsorbed by an adsorbent bed under specific conditions and can
is the amount of adsorbate that is adsorbed. be expressed based on this model as [36,37]:
KLDF is a lumped mass transfer coefficient, as a measure for defining ( ( )n )
how easily vapor moves from the adsorbent surface to the particle at the XEQ = X0 exp − B
T
− 1 (3)
inner points and is defined as [34]: Tsat
( )
15
KLDF = 2 Do exp −
Ea
(2) where n and B are Dubinin-Astakhov (D-A) fitting coefficients and X0 is
rp RT the maximum adsorption capacity for a zeolite 13X-water pair [38].

where rp is the zeolite particle radius, Do is the reference diffusivity, and

3.2. Vapor transport in bed
Ea represents characteristic energy.
The adsorption equilibrium model is utilized to describe the physical
The mass balance equation for the transport of adsorbate gas in the
process of water vapor adsorption in the bed and to link between the
porous domain can be written as [8]:
local circumstances, commonly the temperature and vapor pressure, and
the equilibrium water uptake. The model adopted in this study to ∂C
εeff − Deff ∇2 C + ∇(C.u) = Rs (4)
calculate the equilibrium loading is Dubinin's theory [35]. The basic ∂t
assumption for using this theory is the pore size which should be <2 nm.
where Deff is the physical gas diffusivity in the porous domain, C is the

5
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 4. The quality of computational meshes

Fig. 5. Validation of the model with the experiments conducted by Wu et al. [48].

6
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Çağlar. [31]

Present Study
Fig. 6. Validation of the model for adsorbent bed temperature with the work of Çağlar [31].

water vapor concentration, ρs is the adsorbent density, Mw the water Under the very low-pressure conditions in this system, the water
vapor molar mass, and εeff is the effective porosity and is defined as: vapor can be considered as an ideal gas, and the density can be defined
based on ideal gas relations, so the pressure, P, of the gas can be given as:
εeff = εbed + εp (1 − εbed ) (5)
P = cRT (7)
where εp and εbed are the particle and bed porosity of zeolite.
In Eq. (4), Rs is the reaction term and can be defined as: where T is the temperature and c the water vapor concentration.
( ) The transport of water vapor in the porous media is governed by both
1 − εeff ρs ∂X diffusion and advection. The pressure difference causes the bulk trans­
Rs = − (6)
Mw ∂t port of water vapor. The vapor velocity in axial and radial directions is

7
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Table 4 ( ) ( ) ( )
Design and result of CCD analysis ρC p eff
= εeff ρf Cpf + 1 − εeff ρs Cps + 1 − εeff ρs Cl X (11)

Case Fin length Corresponding width of Angle Average outlet

(mm) the fin (mm) (◦ ) temperature of HTF
Qs is the heat generated from adsorbing water vapor in zeolite particles
(K) and can be calculated by:
1 15 5.88 175 314.817 ( ) ∂X
2 22 4.1 88 318.462 Qs = 1 − εeff ρs |ΔH| (12)
∂t
3 0.8 10.5 88 312.878
4 2 4.45 150 316.651
where ΔH is the reaction enthalpy and is given in Table 2.
5 2 4.45 26 317.258
6 1 8.6 26 313.421 Heat transfer equation can be considered for heat transfer in fins and
7 1 8.6 150 313.366 heat transfer tube [44,45]:
8 15 5.88 88 315.521
9 12 7.26 45 314.4755 ∂Tco hti Ati ( ) hs,co As,co
ρco Cp,co − kco ∇2 Tco + Tco − Thtf + (Tco − Ts ) = 0 (13)
10 15 5.88 0.32 313.852 ∂t Vt Vt

where kco , Cp,co , and ρco are thermal conductivity, specific heat capacity,
and density of the copper fins, respectively.
The governing equation for heat transfer fluid domain includes only
convection and conduction and can be written as:
∂Thtf 2 ( )
ρhtf Cp,htf + ρf Cp,f ∇Thtf − khtf ∇2 Thtf + hti Thtf |r=rt − Tco = 0 (14)
∂t rt

3.4. Definition of Initial and boundary conditions

For the adsorption process, the conditions can be described in the

following way:

• An initially uniform distribution is considered for temperature,

pressure, and adsorbed amount in the entire domain (porous media,
heat transfer fluid, and tube):
X = Xi , P = Pi

Tfin (t = 0) = Tco (t = 0) = Ts (t = 0) = Tin

Fig. 7. The change of average temperature of HTF with the length of fin and α. Xi = Xeq (Pi , Ti )

Pi
modeled with the Darcy equation which is valid for the low gas veloc­ ci =
RTi
ities in porous media like the adsorbent bed [10,39]:
κ
u= − ⋅(∇P − ρW g) (8) • A zero temperature gradient is considered at the inlet and outlet of
μ ⃒
∂T
the tube ( f = 0 ∂Tco ⃒z=0,L = 0)
∂z ∂z
where μ is the gas viscosity, and ρW is the gas density. The porous media • Considering the insulation, it can be assumed that the temperature of
permeability κ is defined with the semi-empirical Blake-Kozeny equa­ HTF at the outlet along the axial direction remains constant ( ∂zhtf =
∂T
tion [40,41]:
0).
dp2 ε3bed • For the bed, a zero temperature gradient is considered at z = 0 and z
κ= (9) ∂T ⃒
150 (1 − εbed )2 = L, due to the insulation ( f = 0, ∂Ts ⃒z=0,L = 0).
∂z ∂z
• The outside boundary of the adsorbent bed is considered adiabatic
because of the low vapor pressure and the negligible heat transfer
⃒
3.3. Heat transfer (∂∂Tr ⃒r=rb = 0).
• The pressure gradient in the direction normal to the plane is set to
Three different domains of adsorbent bed, heat transfer fluid (HTF), zero where the adsorbent boundaries are enclosed by the tube and
fins and heat transfer tube (HTT) are taken into consideration. The heat fins (∂∂Pn = 0).
transfer equations are written for these domains as follows. The gov­ • The vapor in the boundaries of the adsorbent bed which are open to
erning equation for the porous zeolite is written as [42,43]: the evaporator is at the evaporator pressure (P|r=rt = Pevap ).
( ) ∂Ts hs,co As,co
ρCp + ρf Cpf u.∇Ts − keff ∇2 Ts + (Ts − Tco ) = Qs (10)
eff ∂t Vs 3.5. Response surface methodology

where Cpf and ρf are specific heat capacity and density of water vapor, The design optimization applying response surface methodology
( )
and ρCp eff is the effective volumetric heat capacity and is given by: (RSM) has been used widely in both industry and research, with the aim

8
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 8. Variation of temperature and water uptake at 50, 100, and 250 min for cases (a) without fin (b) with straight fin, and (c) with optimized Y-shape fin

9
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

of reducing the number of experiments and at the same time, realizing of all equations was done using the quadratic second-order method.
enough data for understanding the influence of under-study parameters Although the maximum time step is considered equal to 1 s, the solver
on the desired output [46,47]. To better study the effect of branch angle set a very small value for the time step at the initial step of convergence.
and length of the fins on the adsorption process, the simulation exper­
iment has been conducted according to RSM based on central composite 4.1. Meshing and grid independence test
design (CCD) in this study, applying the following steps:
To decrease the error of the calculations, a very decreased grid size is
(1) Selection of the response: The purpose of the present study is to considered close to the interfaces of the porous medium and solid. The
reach the highest possible average outlet temperature at the grid-independent study is conducted on the case with the fin length of
outlet of HTF (Tave,out ) considering different fin designs. 0.015 m and the angle between branches of 105◦ to confirm that the
(2) Choosing variables and assigning codes to them: The parameters obtained solutions are independent of the grid resolution. Three
which influence the response, are defined. In this study, the fin different grid sizes of 22,950, 54,879, and 87,606 triangular elements
geometrical parameters, length (L), ranging from 10 to 20 mm, are considered for simulations, and obtained water loading amounts
and angle between branches (α), ranging from 0◦ to 175◦ are were compared. The results obtained for a very fine mesh with 87,606
considered for the study. elements did not exceed 0.02 % of the results of the case with 54,879
(3) Experimental design development elements. The independent verification of the grid is shown in Fig. 3.
(4) Regression analysis Finally, the mesh with 54,879 elements was used for calculation. The
(5) A quadratic polynomial Formation i.e. response development mesh quality is depicted in Fig. 4. The minimum and maximum element
size of the used grid is 0.0000232 m and 0.00162 m, respectively, so the
4. Numerical method grid is sufficiently dense to get solutions that are accurate and inde­
pendent from the mesh size.
The available finite element method (FEM) based CFD solver,
COMSOL Multiphysics 6.1 is utilized to solve the coupled mass and heat 4.2. Validation
transfer equations. User-defined functions are written for programming
the reaction kinetics and thermophysical properties expressions. The To confirm the accuracy of the current simulation, the model was
available interfaces of ‘heat-transfer-in-porous-media’ and ‘transport-of- validated through the comparison with the results of previous experi­
diluted-species-in-porous-media’ are used for programming the equa­ mental works and also numerical simulations in the literature. The
tions in COMSOL. The computational grid is generated by discretizing developed model has been validated for the desorption process in a
three separated domains of heat transfer fluid (HTF), heat transfer tube closed system with experiments of Wu et al. [48], on a system including
(HTT), and adsorbent bed (porous domain) in extremely small elements, a cylindrical silica gel bed and a water jacket around the bed, acting as a
aiming to get high precision of the numerical results. The discretization heat source. The cylindrical bed had inner and outer radii of 0.01 and

Fig. 9. The effect of the fin type on the outlet temperature of HTF

10
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

0.11 m, respectively. The heat exchange with the bed is through the hot cases is depicted in Fig. 8. As can be observed, the temperature is lower
water flows by a water jacket. A condenser with a pressure of 2337 Pa is around the fins, and the amount of water uptake increases in these areas.
connected to the inner surface of the silica gel. The initial temperature It can be explained by better heat exchange between the cold HTF and
and pressure were 20 ◦ C and 1000 Pa, respectively. The temperature of bed, so the bed temperature decreases significantly in these areas, and
the cooling and heating water was 20 ◦ C and 80 ◦ C, respectively. Silica based on Dubinin-Astakhov-formula, Eq. (3), decreasing the adsorption
gel is dehydrated through the transportation of heat in the bed and bed temperature leads to augmenting the adsorption capacity of zeolite,
thermal energy is stored. The results of the simulation have been so a higher amount of water can be adsorbed. On the other hand,
compared with the data from experiments considering the temperature enhanced heat generation and better heat transfer resulting from opti­
at the radius of 100 mm, as shown in Fig. 5. mum fin design lead to reaching higher outlet temperatures for HTF,
Further, the numerical model has been also validated with the study which is the most important objective in heat storage systems. The
by Çağlar [31] on a closed sorption system. The results are presented in outputs of the numerical investigation for the outlet temperature of HTF
Fig. 6. for Cw , Csf and Copt are given in Fig. 9, which shows that the outlet
As can be observed in Figs. 5 and 6, there is a good agreement be­ temperature of HTF increases significantly using optimized Y-shaped
tween the results from the present developed model and that from fins. Average temperature lift of HTF for Cw , Csf and Copt is 5.95, 9.92,
literature, showing the accuracy of the developed model for further and 14.5 K, respectively, showing a 46 % improvement for the optimized
numerical calculations. case compared with the case with straight fins, and 143.5 % compared
with the case without fins. An enhanced heat transfer in the optimized
5. Results and discussion case leads to not only higher energy transfer to HTF from the zeolite area
but also decreases the bed temperature and augments the adsorption
To study the effect of using Y-shaped fins on the performance of the rate, which in turn augments the heat generation in the adsorption
system, numerical investigations have been performed. Several simu­
lations have been conducted to evaluate the effect of fin length and the
angle between branches on the system performance. Based on the ob­
tained results, the optimal geometry has been proposed for the consid­
ered parameters range. Moreover, the influence of fin length and
numbers, fin type, and angle between branches and evaporator pressure
on the adsorption process has been investigated and evaluated consid­
ering the water uptake of zeolite, bed temperature, outlet temperature of
HTF, and power output. The power output has been defined as the en­
ergy transferred to the HTF per unit of surface area and unit of time.
Further, the performance of the optimized case has been compared with
the case with a straight fin and without fin for both the adsorption and
desorption processes.

5.1. RSM optimization of the geometry

The influence of fin length and the angle between branches on the
performance of the system has been studied utilizing RSM. With
considering these two parameters as independent variables and the
average outlet temperature of HTF as evaluation index, simulations have
been conducted. The results of conducted simulations are shown in
Table 4. It should be mentioned that the pressure of 3 Pa and the initial
temperature of 30 ◦ C are taken for all cases. The evaporator temperature
is considered to be 15 ◦ C, which corresponds to a pressure of 2334 Pa.
According to the data in Table 4, the obtained equation for the
average outlet temperature of HTF from CCD analysis is:
Tout ave = 309.191 ( ) + 0.02186(α) − 0.445231(Lα)
( ) + 348.6953(L) (15)
+2199.6403 L2 − 9.4394 α2

where 0◦ < α < 175◦ , 10 (mm) < L < 20 (mm). The influence of these
parameters on the HTF outlet temperature, based on RSM is presented in
Fig. 7. As can be observed, there is a direct relationship between the
length of the fin and the outlet temperature. That also shows the
importance of the length compared to the width of the used fin, since the
longer length corresponds to a smaller width. On the other hand, with
increasing α, first, the outlet temperature increases but then decreases.
Based on RSM analysis, When the angle is 58◦ and the length is 0.02 m,
the outlet temperature reaches its maximum, 317.5 K.

5.2. The influence of the fin type

Three different cases have been considered for investigations:

without fin (Cw ), with straight fins (Csf ) and with optimized Y-shaped Fig. 10. Variation of (a) water uptake and (b) output power for the case
fins (Copt ). The variation of temperature and water uptake for different without fin, with straight fin, and with Y-shaped fin

11
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 11. The effect of fin numbers on average outlet temperature of HTF.

process, reaching higher outlet temperature in HTF area. provide more heat transfer areas between the adsorption bed and HTF
The amount of adsorbed water and output power at the tube surface area, leading to more efficient heat transfer. So the heat generated from
in contact with HTF for the cases of Cw , Csf , and Copt is presented in the adsorption process can be better absorbed by HTF and the bed
Fig. 10. The equilibrium uptake is much higher than the actual amount temperature decreases, causing higher adsorption capacity and heat
of adsorbate in the bed at the beginning of the process, so based on the generation in the zeolite bed, leading to higher output power.
LDF model, Eq. (14), the water uptake is high, leading to very high The effect of fin numbers on changes in temperature as well as
output power. The sharp increase in output power can be also explained adsorbed water with time is illustrated in Fig. 13. The better heat ex­
by this model. Over time and with adsorbing water, the difference be­ change and higher water uptake with increasing fin numbers lead to a
tween the equilibrium loading and actual uptake reduces, leading to considerable increase in output power, as illustrated in Fig. 12.
decreasing the adsorption rate of water. The higher amount of power
output in Copt , Fig. 10b, is due to the higher amount of adsorbed water by
the adsorbent bed. The average amount of output power in the time 5.4. The influence of the length and branch angles of fins
period of 800 min for cases of Cw , Csf , and Copt is 668.6 W/m2 , 1115.7
W/m2 , and 2265.4 W/m2 , respectively. The numerical investigations have been performed for different fin
lengths between 0.01 and 0.02 m. It should be noted, that since the
volume of the fins is kept constant for all cases, the fins with a longer
5.3. The effect of fin numbers length have shorter widths. According to the outputs of simulations as
well as the RSM analysis, the length of the fin has the most considerable
To evaluate the influence of fin numbers on the performance of the effect on the system performance. With increasing the fin length, the
system, different fin numbers of 2, 4, 6, 8, and 10 have been considered. heat exchange between the bed improves, leading to a higher outlet
The length, width and angle, and boundary conditions for all cases are temperature of HTF. At the same time, better heat exchange causes
the same as in the optimal case defined based on RSM analysis. The decreasing the bed temperature through the adsorption process, leading
results of an investigation on the temperature, water uptake, and output to augmenting the adsorption capacity of zeolite. As a result, the amount
power at different time intervals are presented in Figs. 11 and 12. It can of adsorbed water and the adsorption rate of the process increases,
be clearly observed in Fig. 10, that with increasing fin numbers, the which also increases the heat generation in the process.
water uptake and average outlet temperature of HTF increase. More fins The angle between two branches of the fin is referred to as the branch

12
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

determine to which degree the storage system performance is influenced

by this pressure, simulations are performed for three different evapo­
rator temperatures of 278.15 K, 283.15 K, 293.15 K, and 298.15 K
corresponding to the pressure of 837, 1230, 2334, and 3176 Pa. The
influence of this pressure on outlet temperature variations and water
uptake is presented in Fig. 16. Based on the results, the amount of water
uptake increases with increasing the evaporator pressure, leading to a
greater temperature in the adsorption bed, and outlet temperature of
HTF, as can be observed in Fig. 17. This can be explained based on the
reaction kinetics. The adsorption potential increases with the increase of
evaporator pressure, leading to augmenting water vapor adsorption rate
and increasing bed temperature. Dubinin-Astakhov (D–A) equation says
that by increasing the pressure of adsorbate and decreasing the tem­
perature of adsorbent, the adsorption capacity of zeolite augments.

5.6. The influence of the optimum design on the desorption process and
the efficiency of the reactor

In order to study the effect of adding fins to the bed on the desorption
process, simulations were performed for two cases without fins and with
Y-shaped fins (optimum case). The parameters used for developing the
numerical model of the desorption process are given in Table 5.
Considering the boundary conditions, the vapor in the boundaries of
the bed which are open to the condenser is considered at the condenser
pressure (P|r=rt = PCon ) and X(t = 0, r, z) = X0 = 0.26. The rest of the
initial and operating conditions and characteristics are considered the
same as for the adsorption process.
The results of the simulation in terms of the average temperature of
the adsorbent bed over time are reported in Fig. 18. As can be observed,
the average temperature of the zeolite bed has been significantly
increased using the optimum shape of Y-shaped fins, resulting in
improved performance of the desorption process. The average temper­
ature for the time period of 800 min is 396 K and 354 K for the optimum
case and the case without fins, respectively.

5.7. The influence of the optimum design on the efficiency of the reactor

The desorption process involves the release or removal of the

adsorbate molecules from the sorbent structure and is an endothermic
reaction. During this process, the heat is stored in the zeolite bed due to
the thermochemical dissociation between water and zeolite. It can be
Fig. 12. The effect of fin numbers on (a) water uptake and (b) output power. achieved by applying heat from an external heat source. A part of pro­
vided heat is stored in the form of chemical energy. The efficiency of this
process can be defined as the enthalpy change during the desorption
angle. With increasing this angle from 0◦ at the constant length of the reaction divided by the energy provided by HTF:
fin, the heat transfer process improves, because there is more contact ( ∫∫ )
Qs dvdt Des
area between the bed and fin, until an optimized angle. After this point, ηDes = ∫ t ( ) (16)
ṁCp,htf TDes,in − TDes,out dt
the performance decreases, which can be explained by considering the 0

fact that the water vapor cannot easily penetrate into the areas under the
where ṁ is the mass flow rate of the HTF, Cp,htf is the specific heat ca­
secondary branch, which can negatively affect the adsorption process.
pacity of the HTF, TDes, in and TDes, out are the inlet and outlet temper­
However, the influence of the branch angle on the adsorption process
atures of the HTF.
here is not too strong. The temperature and water uptake contours and
During adsorption, water molecules in the gas phase adhere to the
also output power for three different cases with the same fin length of 15
surface of the zeolite material. As a result, the enthalpy of adsorption is
mm but different branch angles of 10, 58, and 175◦ are presented in
released and extracted by the same HTF cycle as in the desorption phase
Figs. 14 and 15.
via the reactor heat exchanger. The efficiency of the adsorption process
can be quantified by the ratio of the useful energy absorbed by HTF to
5.5. The influence of pressure in the evaporator the heat released during the reaction [49]:
∫t ( )
One important factor influencing the efficiency of the adsorption ṁCp,htf TAd,out − TAd,in dt
ηAd = 0 ( ∫∫ ) (17)
process, and so the reachable outlet temperature is the concentration of Qs dvdt Ad
incoming water vapor flowing from the evaporator into the reactor. The
concentration depends on the saturation pressure of the water vapor TAd, out and TAd, in are the outlet and inlet temperatures of the HTF.
flow, which itself depends on the temperature of the evaporator. To The efficiency of the reactor during the desorption process has been

13
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 13. The effect of fin numbers on the temperature as well as water uptake variations with time for cases (a) with 2 fins, (a) with 4 fins, and (a) with 10 fins

14
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 14. The effect of the branch angles on the variation of water uptake and temperature with time for cases with Y-shaped fins with (a) L = 15 mm, α = 10◦ , (b) L
= 15 mm, α = 58◦ , and (c) L = 15 mm, α = 175◦

15
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 15. The influence of branch angle on the output power

calculated based on Eq. (16) and defined to be 69.2 %, 65.5 %, and 37.7
% for the optimum Y-shaped fin, straight fin, and no fin cases, respec­
tively. The efficiency of the adsorption cycle, based on Eq. (17), has been
also calculated to be 75.3 %, 57.2 %, and 42.7 % for the optimum Y-
shaped fin, straight fin, and no fin cases, respectively.

6. Conclusions

In the current study, the transient coupled heat and mass transfer in a
zeolite bed containing straight and Y-shaped fins have been modeled,
aiming to overcome the limitation of heat transfer in the bed. A para­
metric study is performed to study the influence of fin geometrical pa­
rameters, length, and branch angles, on the adsorption process
performance. The average outlet temperature of HTF has been consid­
ered as the criterion, and the optimized geometry of the fin has been
defined by applying RSM analysis. Based on this analysis, the effect of fin
length is much higher than the angle between the branches. For the
optimal design, an increase of 4.6 K and 8.7 K in the average outlet
temperature is observed compared to the case with a straight fin and the
case without fin, respectively. The average output power of 2265.4 W/
m2 is calculated for the optimum design, which shows 103 % and 238.8
% improvement compared to the case with straight fins and without fin,
respectively. The optimized case with 6 fins has been taken into
consideration and the influence of different parameters such as evapo­
rator pressure and fin numbers has been investigated in the optimized
case considering the output power, water uptake, and HTF outlet tem­
perature. Based on the results, the power output and HTF outlet tem­
perature enhance significantly with the increase of evaporator pressure,
as a result of augmenting water uptake. The effect of fin numbers has
been also studied and the best performance has been observed for the
case with 8 fins. Investigating the performance of the system in two
different designs without fins and including Y-shaped fins during the
desorption process reveals a significant increase in bed temperature by
adding optimized Y-shaped fins. The average temperature in the zeolite
bed is reported to be 345 K in the case without fins and 396 K in the
optimum case. The efficiency of the reactor for the optimum design has
been defined as 69.2 and 75.3 % for the desorption and adsorption
Fig. 16. Variation of (a) temperature and (b) output power for different processes, respectively.
evaporator pressures.

16
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Fig. 17. Variation of temperature and water uptake at 50, 100, and 250 min for the optimized case at Pevap = (a) 873.35 Pa, (b) 1230 Pa, and (c) 3170 Pa.

17
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

Table 5 [6] U. Tesio, E. Guelpa, C. Ortiz, R. Chacartegui, V. Verda, Optimized synthesis/design

Parameters used for the desorption process model of the carbonator side for direct integration of thermochemical energy storage in
small size concentrated solar power, Energy Conversion and Management: X 4
Parameter Value Unit Description (2019) 100025.
[7] L. Scapino, C. De Servi, H.A. Zondag, J. Diriken, C.C.M. Rindt, A. Sciacovelli,
PCon 4247 Pa Condenser pressure Techno-economic optimization of an energy system with sorption thermal energy
Tinit 303.15 K Initial temperature storage in different energy markets, Appl. Energy 258 (2020) 114063.
Thot 453 K Hot HTF inlet temperature [8] A. Çağlar, The effect of fin design parameters on the heat transfer enhancement in
the adsorbent bed of a thermal wave cycle, Appl. Therm. Eng. 104 (2016) 386–393.
[9] S. Mitra, M. Muttakin, K. Thu, B.B. Saha, Study on the influence of adsorbent
particle size and heat exchanger aspect ratio on dynamic adsorption
characteristics, Appl. Therm. Eng. 133 (2018) 764–773.
[10] K. Kant, A. Shukla, D.M.J. Smeulders, C.C.M. Rindt, Analysis and optimization of
the closed-adsorption heat storage bed performance, J. Energy Storage 32 (2020)
101896.
[11] B. Golparvar, H. Niazmand, A. Sharafian, A.A. Hosseini, Optimum fin spacing of
finned tube adsorber bed heat exchangers in an exhaust gas-driven adsorption
cooling system, Appl. Energy 232 (2018) 504–516.
[12] K. Kant, R. Pitchumani, Analysis of a novel constructal fin tree embedded
thermochemical energy storage for buildings applications, Energ. Conver. Manage.
258 (2022) 115542.
[13] C. Reichl, D. Lager, G. Englmair, B. Zettl, M. Popovac, Fluid dynamics simulations
for an open-sorption heat storage drum reactor based on thermophysical kinetics
and experimental observations, Appl. Therm. Eng. 107 (2016) 994–1007.
[14] A. Kurniawan, A. Rachmat, CFD Simulation of Silica Gel as an Adsorbent on Finned
Tube Adsorbent Bed, EDP Sciences, p. 01014.
[15] A. Crespo, A. Frazzica, C. Fernández, Á. de Gracia, Optimizing the discharge
process of a seasonal sorption storage system by means of design and control
approach, Journal of Energy Storage 60 (2023) 106652.
[16] R.H. Mohammed, O. Mesalhy, M.L. Elsayed, L.C. Chow, Performance evaluation of
a new modular packed bed for adsorption cooling systems, Appl. Therm. Eng. 136
(2018) 293–300.
[17] R. Salgado-Pizarro, A. Calderón, A. Svobodova-Sedlackova, A.I. Fernández,
C. Barreneche, The relevance of thermochemical energy storage in the last two
decades: the analysis of research evolution, J. Energy Storage 51 (2022) 104377.
[18] B. Mette, Experimentelle und numerische Untersuchungen zur Reaktionsführung
thermochemischer Energiespeicher, 2014.
[19] P. Tatsidjodoung, N. Le Pierrès, L. Luo, A review of potential materials for thermal
energy storage in building applications, Renew. Sustain. Energy Rev. 18 (2013)
327–349.
Fig. 18. The variation of bed temperature for the optimum case and the case [20] X. Zheng, T.S. Ge, R.Z. Wang, Recent progress on desiccant materials for solid
without fin during the desorption process desiccant cooling systems, Energy 74 (2014) 280–294.
[21] D. Jähnig, R. Hausner, W. Wagner, C. Isaksson, Thermo-Chemical Storage for Solar
Space Heating in a Single-Family House, Proceeding of Ecostock, New Jersey,
CRediT authorship contribution statement 2006.
[22] A.A. Askalany, M. Salem, I.M. Ismail, A.H.H. Ali, M.G. Morsy, A review on
adsorption cooling systems with adsorbent carbon, Renew. Sustain. Energy Rev. 16
Elham Abohamzeh: Conceptualization, Data curation, Investiga­ (1) (2012) 493–500.
tion, Methodology, Software, Validation, Visualization, Writing – orig­ [23] L. Scapino, H.A. Zondag, J. Van Bael, J. Diriken, C.C.M. Rindt, Sorption heat
inal draft, Writing – review & editing. Seyed Ehsan Hosseinizadeh: storage for long-term low-temperature applications: a review on the advancements
at material and prototype scale, Appl. Energy 190 (2017) 920–948.
Conceptualization, Investigation, Methodology, Software, Validation, [24] B.H. Thomas Herzog, Jochen Jänchen, Udo Hellwig, New innovative molecular
Visualization. Georg Frey: Conceptualization, Investigation, Method­ sieves in energy engineering, 25. Deutsche Zeolith-Tagung, Hamburg, Germany.
ology, Project administration, Resources, Supervision, Writing – original [25] A.I. Khuri, S. Mukhopadhyay, Response surface methodology, Wiley
Interdisciplinary Reviews: Computational Statistics 2 (2) (2010) 128–149.
draft, Writing – review & editing. [26] C. Liu, H. Yang, Multi-objective optimization of a concrete thermal energy storage
system based on response surface methodology, Appl. Therm. Eng. 202 (2022)
117847.
Declaration of competing interest [27] F. Ahmed, M.S. Abdul Aziz, M.R.R. Mohd Arif Zainol, K.C. Yee, F. Shaik, D.S.
Che Halin, M.A.A. Mohd Salleh, M. Kheimi, Design, modelling and optimization of
a novel concentrated solar powered (CSP) flash desalination system involving
The authors declare that they have no known competing financial direct heating and pressure modulation using response surface methodology
interests or personal relationships that could have appeared to influence (RSM), Sustainability 14 (18) (2022) 11558.
the work reported in this paper. [28] L. Sun, C.-L. Zhang, Evaluation of elliptical finned-tube heat exchanger
performance using CFD and response surface methodology, Int. J. Therm. Sci. 75
(2014) 45–53.
Data availability [29] H. Alshihmani, M.-J. Maghrebi, M. Sardarabadi, Thermal performance prediction
of a phase change material based heat-sink cooling system for a printed circuit
board, using response surface method, J. Energy Storage 55 (2022) 105499.
Data will be made available on request.
[30] R.-J. Clark, A. Mehrabadi, M. Farid, State of the art on salt hydrate thermochemical
energy storage systems for use in building applications, J. Energy Storage 27
References (2020) 101145.
[31] A. Çağlar, Design and Experimental Testing of an Adsorbent Bed for a Thermal
Wave Adsorption Cooling Cycle, 2012.
[1] P. Huovila, M. Ala-Juusela, L. Melchert, S. Pouffary, C.-C. Cheng, D. Ürge-Vorsatz,
[32] İ. Solmuş, An Experimental Study on the Performance of an Adsorption Cooling
S. Koeppel, N. Svenningsen, P. Graham, Buildings and Climate Change: Summary
System and the Numerical Analysis of its Adsorbent Bed, 2011.
for Decision-Makers, 2009.
[33] F.W. John Thomas, B. Crittenden, Adsorption Technology and Design,
[2] L. Pérez-Lombard, J. Ortiz, C. Pout, A review on buildings energy consumption
Butterworth-Heinemann, 1998.
information, Energ. Buildings 40 (3) (2008) 394–398.
[34] E. Glueckauf, Theory of chromatography. Part 10.—Formulæ for diffusion into
[3] G. Energy, CO2 Status Report, IEA (International Energy Agency), Paris, France,
spheres and their application to chromatography, Trans. Faraday Soc. 51 (1955)
2019.
1540–1551.
[4] T.R.S. Gbenou, A. Fopah-Lele, K. Wang, Recent status and prospects on
[35] B.P. Bering, M.M. Dubinin, V.V. Serpinsky, Theory of volume filling for vapor
thermochemical heat storage processes and applications, Entropy 23 (8) (2021)
adsorption, J. Colloid Interface Sci. 21 (4) (1966) 378–393.
953.
[36] R.E. Critoph, H.L. Turner, Performance of Ammonia-Activated Carbon and
[5] R. Bravo, C. Ortiz, R. Chacartegui, D. Friedrich, Hybrid solar power plant with
Ammonia Zeolite Heat Pump Adsorption Cycles, pp. 202–11.
thermochemical energy storage: a multi-objective operational optimisation, Energ.
Conver. Manage. 205 (2020) 112421.

18
E. Abohamzeh et al. Journal of Energy Storage 84 (2024) 110803

[37] İ. Solmuş, C. Yamalı, B. Kaftanoğlu, D. Baker, A. Çağlar, Adsorption properties of a [44] A. Çağlar, C. Yamalı, D.K. Baker, Two dimensional transient coupled analysis of a
natural zeolite–water pair for use in adsorption cooling cycles, Appl. Energy 87 (6) finned tube adsorbent bed for a thermal wave cycle, Int. J. Therm. Sci. 73 (2013)
(2010) 2062–2067. 58–68.
[38] L.W. Wang, R.Z. Wang, R.G. Oliveira, A review on adsorption working pairs for [45] F.P. Incropera, D.P. DeWitt, T.L. Bergman, A.S. Lavine, Fundamentals of Heat and
refrigeration, Renew. Sustain. Energy Rev. 13 (3) (2009) 518–534. Mass Transfer, Wiley, New York, 1996.
[39] C. Latrille, A. Zoia, Estimating apparent diffusion coefficient and tortuosity in [46] A. Coppedè, S. Gaggero, G. Vernengo, D. Villa, Hydrodynamic shape optimization
packed sand columns by tracers experiments, J. Porous Media 14 (6) (2011). by high fidelity CFD solver and Gaussian process based response surface method,
[40] M.J. MacDonald, C.F. Chu, P.P. Guilloit, K.M. Ng, A generalized Blake-Kozeny Appl. Ocean Res. 90 (2019) 101841.
equation for multisized spherical particles, AICHE J. 37 (10) (1991) 1583–1588. [47] H. Jahangir, J.A. Esfahani, M. Pourali, K.C. Kim, Parameter study of a porous solar-
[41] A.D. Vdi-Wärmeatlas, 23: Wärmeleitfähigkeit der Flüssigkeiten in W/mK. 7. based propane steam reformer using computational fluid dynamics and response
erweiterte Auflage, VDI-Verlag, Düsseldorf, 1994. surface methodology, Int. J. Hydrogen Energy 47 (86) (2022) 36465–36481.
[42] L.M. Sun, N.B. Amar, F. Meunier, Numerical study on coupled heat and mass [48] J.W. Wu, M.J. Biggs, E.J. Hu, Dynamic model for the optimisation of adsorption-
transfers in an absorber with external fluid heating, Heat Recovery Systems and based desalination processes, Appl. Therm. Eng. 66 (1–2) (2014) 464–473.
CHP 15 (1) (1995) 19–29. [49] M. Abou Elfadil, Investigations and Technical Development of Adsorption Thermal
[43] M.W. Ellis, An Evaluation of the Effect of Adsorbent Properties on the Performance Energy Storage Systems with Simulation and Different Control Strategies, 2021.
of a Solid Sorption Heat Pump, Georgia Institute of Technology, 1996.

19