Journal of Energy Storage 28 (2020) 101209

Contents lists available at ScienceDirect

Journal of Energy Storage

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

A CFD modeling and investigation of a packed bed of high temperature T

phase change materials (PCMs) with different layer configurations
Farshad Mohammadnejada, Siamak Hossainpourb,⁎
a
Faculty of Mechanical Engineering, Sahand University of Technology, Tabriz, Iran
b
Faculty of Mechanical Engineering, Sahand University of Technology, Tabriz 51325-1996, Iran

A RT ICLE INFO ABSTRACT

Keywords: In recent years, one of the most important issues is optimization of thermal energy-storing technologies. Using
Phase change materials (PCM) phase change materials (PCMs) with specific phase change temperatures allows us to store latent heat as thermal
Heat transfer energy. This is one of the most efficient ways to store thermal energy. The objective of this paper is to investigate
Energy storage tanks the discharge performance of a packed bed filled with multi layers of high temperature encapsulated PCMs
Packed bed
through a CFD modeling via Comsol multi-physics software. Results of simulation are compared to results of the
Porous media
Discharge process
experimental researches of the literature and there is good agreement between them. The influence of each
layer's height and porosity along with different configurations of the PCMs according to their phase change
temperatures (ascending and descending in flow direction) are analyzed. It is observed that increasing the inlet
velocity of heat transfer fluid (HTF) and increasing the porosity of layers of PCM in the packed bed have similar
outcomes and makes the packed bed deplete more rapidly. Some new configurations are suggested and opti-
mized regarding Δ tutl, which is defined as the ratio between duration of receiving energy with desired tem-
perature to total discharge time of the packed bed, during the discharge process. Results indicate that by setting
the layer heights of the PCMs in a descending order and also having the porosity of each layer to decrease in the
flow direction, results with 29.2% better performance can be obtained. Optimization of Δtutl can be a defining
factor in designing of packed beds and also a concentrating solar power (CSP) plant.

1. Introduction better utilize the energy stored with regards to time and amount of
energy needed for each application.
The utilization of the renewable energy resources has attracted The PCM-based heat storage in the packed bed is one of the most
many efforts in recent years due to the continuous increase in the level suitable storage unitsInvalid source specified [7]. A packed bed is a
of greenhouse gas emissions [1]. Solar energy is an important source of volume of porous media obtained by packing particles of selected ma-
renewable energy. Its implementation has some challenges such as a terial into a containerInvalid source specified [8]. Thermocline storage
gap between demand and supply because of the fluctuating nature [2]. concept has been considered for more than a decade as a possible so-
Thermal energy storage is a crucial solution to fill this gap and help to lution to reduce the huge cost of the storage system in concentrated
produce power steadily by solar energy [3, 4]. solar power (CSP) plants [9]. However, one of the drawbacks of this
Sensible heat storage and latent heat storage (LHS) are the basic concept is the decrease in its performance throughout the time. Galione
types of thermal energy storage techniques. LHS technique provides a et al. [10] considered a new thermocline-like storage concept, which
high energy storage density and has the capacity to store heat as latent aims at circumventing this issue. The proposed numerical concept
heat of fusion at a constant temperature. Materials used for latent heat consists of a storage tank filled with a combination of solid material and
thermal energy storage are known as phase change materials (PCMs) encapsulated PCMs, forming a multi-layered packed bed, with molten
and may undergo solid-solid, solid-liquid and liquid-gas phase trans- salt as the heat transfer fluid. The analysis shows that the multi-layered
formations [5]. PCMs require their upper and lower phase transition solid-PCM concept is a promising alternative for thermal storage in CSP
temperatures to be within the operational temperature range for a given plants.
application [6]. Due to abundance of solar energy and time required for Ismail and Stuginsky [11] presented comparative numerical in-
charging process, it is important to study discharge process in order to vestigation on packed bed thermal models suitable for sensible and

⁎
Corresponding author.
E-mail addresses: [email protected] (F. Mohammadnejad), [email protected] (S. Hossainpour).

https://doi.org/10.1016/j.est.2020.101209
Received 22 September 2019; Received in revised form 8 January 2020; Accepted 8 January 2020
Available online 15 January 2020
2352-152X/ © 2020 Elsevier Ltd. All rights reserved.
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Nomenclature Uf Inlet velocity (m/s) LHS Latent Heat Storage

wt% Weight percent
Symbols Δtutl Total energy utilization
H Cylinder height
Cp Specific heat capacity(J/Kg.K) μ Dynamic viscosity (Pa.s) R Cylinder radius
D Diameter(mm) ρ Density (Kg/m3)
K Thermal conductivity(J/Kg) Greek letters
L Latent heat CFD Computational Fluid Dynamics
P Pressure (Pa) CSP Concentrating Solar Power α Phase change function
T Temperature (∘C) HTF Heat Transfer Fluid θ 1–α
t time (sec) PCM Phase Change Material

latent heat thermal storage systems. Ismail and Henrı́quez [12] pre- dimensional, dispersion-concentric model to investigate the dynamic
sented a numerical model to simulate a storage system composed of performance of a high temperature packed bed consisting of rocks only
spherical capsules filled with PCM and packed inside a cylindrical tank. and PCM/rocks combination. Efficiency analysis was conducted and it
They studied the influence of the geometrical and operational para- was found that suitable cut-off temperature can improve charging/
meters of the system on the charging and discharging processes. discharging efficiency.
Geissbuhler et al. [13] used an experimental-numerical approach to The use of fins, PCM embedded in porous matrices, dispersion of
study combined sensible/latent heat storage based on placing a limited highly conductive particles within the PCM, and employing multiple
amount of steel encapsulated AlSi12 on top of a packed bed of rocks. PCMs have been considered as additional ways to improve the perfor-
Then a simulation model was validated accordingly. The simulations mance of the LHS systemsInvalid source specified [23]. The perfor-
showed that the combined storage can reduce the material costs for a mance of the LHS system during charging and discharging mainly de-
given maximum outflow temperature drop during discharging. Felix pends on the difference between the HTF temperature and the melting
Regin et al. [14] developed a model for a packed bed latent heat point of PCM. Unlike single PCM utilization, employing multiple PCMs
thermal energy storage using paraffin wax as PCM to predict the can provide nearly a constant temperature difference during melting
thermal behavior of the system in both charging and discharging pro- and solidification processes. This leads to an almost constant heat flux
cesses. Oró et al. [15] studied heat transfer during phase change of the to the PCM or vice versa during the charging and discharging processes,
PCM experimentally and modeled it by Brinkman equation and energy respectivelyInvalid source specified [24]. Using multiple PCMs with
equation as two different mathematical models. The comparison be- different melting temperatures in a cascaded heat storage system is one
tween both models indicated that free convection was not as important of the many methodologies used for improving heat transfer, which is
as forced convection. Wu and Fang [16] performed a numerical in- vastly influenced by the order of PCMs positioned in the packed bed.
vestigation for a solar heat storage system using spherical capsules Packed bed thermal energy storage systems can provide a large amount
filled with meristic acid as PCM. The energy balance of heat transfer of heat transfer area in a relatively small volume.
fluid (HTF) and PCM was developed as mathematical model to calculate Watanabe et al. [25] studied a heat storage module consisted of
the temperatures of PCM and HTF, solid fraction and heat release rate horizontal cylindrical capsules filled with three types of PCMs. The
during the solidification process. Wu et al. [17] investigated the dy- employment of the three types of PCMs improved charging and dis-
namic thermal performance of a molten-salt packed-bed thermal energy charging rates. Watanabe and Kanzawa [26] optimized a LHS system
storage (TES) system using capsules filled with high-temperature phase with PCMs having different melting points on the basis of second law
change material (PCM). A transient two-dimensional dispersion-con- analysis and the optimum melting points distribution was estimated.
centric (D-C) model is modified to account for the phase change process Yang and Zhang [27] studied numerically a new packed bed using a
within capsules in order to determine the temperature distribution and stratified phase change material. The energy and exergy performances
phase change front within each capsule. It was found that the effective were analyzed and compared with those of the packed bed using only
discharging efficiency of the system can be increased by increasing the one kind of PCM. Yang et al. [28] studied charging performance of a
phase change temperature, decreasing the molten-salt inlet velocity or solar storage packed bed using multiple-type PCMs and the results
decreasing the capsule diameter. Wu et al. [18] studied dynamic showed higher energy and exergy transfer efficiency in comparison
characteristics of solar heat storage system with spherical capsules with the traditional single PCM packed bed. Peiró et al. [29] evaluated
packed bed during discharging process. Paraffin was used as heat sto- experimentally the advantages of using the multiple PCM configuration
rage material and the effects of operational parameters on the heat instead of the single PCM in thermal energy storage systems at pilot
released rate were discussed. They found that higher porosities of plant scale. Narasimhan [30] argues that by using multiple PCMs,
packed bed can cause in reducing the heat release capacity. So 0.4–0.5 thermodynamic irreversibility could be minimized and both energy and
was chosen for porosity of the packed bed. Karthikeyan et al. [19] exergy efficiencies would be higher. As it can be inferred from above
carried out the parametric studies on a packed bed storage suitable for literature, there are various parameters effective in increasing the
low temperature air heating applications using the enthalpy based performance of the packed bed storage systems, but too little focus has
numerical model. Increasing heat transfer surface area with small sized been given to the different settings of the packed bed systems and as a
balls, higher temperature difference between HTF inlet and the phase result their influence on the performance of the system are somewhat
change temperature of PCM, and increasing mass flow rate of HTF, unknown.
were known as important factors during the charging process. In the present study, the influence of different alignments as well as
According to the previous literature, different numerical models the importance of various porosity coefficients and layer heights (in the
have been developed to analyze the performance of the thermal energy ascending and descending order) on the total thermal performance of
storage systems. These models can be mainly divided into single phase the system have been analyzed. Subsequently, in order to better utilize
models, Schumman's model, concentric dispersion model, and con- the amount of energy that is stored in a packed bed, the influence of
tinuous solid phase model [20]. Peng et al. [21] investigated the these parameters on the Δtutl factor which is defined as the ratio be-
thermal behavior of a packed bed latent thermal energy storage using tween duration of receiving energy with desired temperature to total
PCM capsules by applying the concentric dispersion model for system discharge time of the packed bed, will be investigated.
analysis. Liao et al. [22] developed a modified transient, one-

2
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

2. Model description T(t = 0s)=400 C (14)

T
2.1. Mathematical model of the PCM packed bed and governing equations (r = 0) = 0
r (15)
The storage unit under study is based on the configuration of a Effect of the HTF inlet velocity is also investigated. Five different
packed bed consisting of a cylindrical tank which is filled with spherical velocities are examined. Velocities are 0.008, 0.016, 0.03, 0.048, 0.064
PCM containers. The schematic of the packed bed thermal storage and 0.08 m/s. The Effect of different porosities is also investigated. In
system is shown in Fig. 1. As it can be seen in Fig. 1, packed bed is filled order to do so, the tank was filled with spheres with porosities of 0.4,
with three types of PCMs. These PCMs are stored in spherical capsules 0.5 and 0.6. And finally, the effects of five different layouts on the
with diameter of 25 mm and porosity of 0.4. performance of the packed bed are studied.
HTF enters from bottom of the tank with temperature of 250∘C and Table 1 demonstrates the main parameters and characteristics of the
velocity of 0.008 m/s and leaves it from the top. A 2-D axisymmetric packed bed thermal storage system that has been investigated.
numerical model has been developed with COMSOL 5.2 platform to
investigate the thermal behavior of the PCMs. The velocity field is 2.2. Material properties
calculated by solving Eqs. (1) and (2) as below:
·(u) = 0 (1) The performance of latent heat storage system during charging or
discharging processes significantly depend on the temperature differ-
(u· ) u = ·[ p + µ ( u ) + ( ·u )T )] + F (2) ence between the inlet temperature of the HTF and the melting tem-
perature of phase change materials. Using multiple PCMs in order of
Heat transfer from HTF takes place by convection and it is calcu-
decreasing melting temperature can provide a relatively constant phase
lated by solving energy Eq. (3) as follows:
change temperature during melting of phase change materials which is
Cp
T
+ Cp u. T= . (k T ) an advantage over single stage heat storage systems that have a de-
t (3) creasing temperature difference in flow direction of the HTF. In this
The energy Eq. (4) is solved in order to calculate heat transfer from study, three PCMs with upper 300∘C melting points are examined in a
wall of the spheres to HTF, which takes place by conduction me- packed bed of thermal energy storage system. The adopted HTF is a
chanism, as below: molten salt (60% wt NaNO3+40% wt KNO3), and its properties can be
calculated by Invalid source specified [31].
T
Cp = . (k T ) (Kgm 3) = 2090
t (4) f 0.636Tf ( C) (16)
Temperature distribution in PCMs is calculated by solving Eq. (5) Cp . f (JKg 1K 1) = 1443 0.172Tf ( C) (17)
and using the phase change function, α, which is an indicator of the
ratio of the amount of the PCM that has experienced phase change in Kf (Wm 1K 1) = 0.443 0.00019Tf ( C) (18)
the discharge process. This method considers that the phase change
occurs in the interval of (Tm Tm/2) and (Tm + Tm/2 ). Until the µ f (Kgm 1s 1) = 0.022714 1.2 × 10 4Tf ( C) + 2.281 × 10 7Tf2 ( C)
temperature is less than (Tm Tm/2 ) the value of function of phase 10T 3 (
change, α, is zero and when the temperature exceeds (Tm + Tm/2 ), its
+ 1.474 × 10 f C) (19)
value is 1. During the phase change, heat conductivity, density and heat Molten salts are the most used materials for thermal energy storage
capacity are also changed. Eqs. (7)–(9) represent these changes: at high temperatures. The most important factor affecting the perfor-
T mance of thermal energy storage (TES) system is the thermal stability of
Cp + Cp u. T= . (k T ) the materials used to store heat, i.e. the temperature interval where
t (5)
they are liquid. The thermal stability of nitrate molten salts allows the
=1 (6)
heat to be stored in an extended range of very high temperatures. The
= + (1 ) nitrate salt (molten salt) is a mixture of 60% by weight sodium nitrate
phase1 phase2 (7)
(NaNO3) and 40% by weight potassium nitrate (KNO3). It is stable in
1 m air and has a low vapor pressureInvalid source specified [31].
Cp = ( phase1 Cp, phase1 + (1 ) phase2 Cp, phase2 ) +L
T (8)

K = Kphase1 + (1 ) Kphase2 (9)

1 (1 ) phase2 phase1
m =
2 phase1 + (1 ) phase 2 (10)
By using Eqs. (1)–(9), the thermal performance of two different
configurations of the three chosen PCMs in the packed bed, which have
been installed in the ascending and descending order of their phase
change temperatures, are investigated. PCM and total volumes for tank
are 1.15m3 and 6.3m3 respectively. Inlet temperature of the HTF is the
250∘C. The geometries of the tank are shown in Fig. 4.
The boundary and initial conditions for HTF:
T(Y, r, 0)=400 C (11)

T(0, r, t)=250 C (12)

T
(Y , R , t ) = 0
r (13)
The boundary and initial conditions for PCM particles: Fig. 1. Schematics of the packed bed filled with PCM spherical capsules.

3
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Table 1 H = 175 mm, respectively, during the charging process.

Main parameters of the system. As it is clear, trends of both curves in the both parts are so identical
Parameters values that the maximum relative difference of them is about 3.5%. The dis-
crepancies observed in between numerical and experimental results can
Bed height (H) 5m be attributed to the fact that in the experimental setup the tank's en-
Bed diameter (D) 1m
tering and exit areas are not perfectly insulated and therefore, some
Capsule diameter (dp) 25 mm
Fluid velocity (uf) 0.008 m/s
heat losses appear in the system and also there are differences between
HTF 60%NaNO3+40%KNO3. actual and simulated heat transfer coefficients. Despite these dis-
PCMs NaNO2, KNO3, KOH crepancies, it seems that numerical results of the present model are in a
Bed temperature range 250–400 °C good agreement with previous experimental data; thus making further
predictions acceptable.

Thermo-physical properties of the PCMs are reported in Table 2.

3. Results and discussion
2.3. Mesh and boundary condition
As it was discussed earlier, different kinds of PCMs can be used to
Due to considerable amount of calculations, required time and increase the heat transfer between HTF and capsules. The heat transfer
rate or generally LHS performance during charging and discharging
memory in the 3-D form of the present study; an axisymmetric 2-D
model has been chosen. For the sake of precision, triangular and rec- processes primarily depends on the temperature difference of HTF and
phase change temperature of capsules. In the case of using only one
tangular elements have been utilized and the mesh sizes vary from 10 to
0.02 mm based on the required accuracy. An extra fine mesh size has kind of capsule, this temperature difference obviously decreases along
the flow direction which decreases the heat transfer rate and results in
been considered for the whole model and for the monitored heights an
extremely fine mesh size has been used so the precision of the calcu- lower performance of the system. While using various kinds of capsules
with different phase change temperatures produces a constant tem-
lations be utterly improved. The assumptions below have been con-
perature difference during charging and discharging processes. That is,
sidered for boundary conditions:
if capsules are configured appropriately. In other words, choosing a

• HTF's temperature and velocity are constant. correct configuration can make an approximately constant heat flow

• Center of the tank is considered to be the symmetric line. along the packed bed and help increase the efficiency of energy storage

• Perimeter surface of the tank is assumed to be thermally insulated. process.

• The distributors are not included in the computational domain and a 3.1. Comparison of two different PCM layouts
uniform flow is imposed at the inlet of the filler region.
• There is no internal heat generation in the bed.
• Radiant heat transfer is negligible. Fig. 4 illustrates placement of three different PCMs in a packed bed.
• PCMs go through three stages of phase change i.e. solid phase, phase Capsules are located in order of their phase change temperatures which
is ascending and descending in case A and B, respectively.
transition and liquid phase.
• The properties of PCMs during phase change are calculated by phase Figs. 5 and 6 depict temperature distributions of HTF and capsules
in different axial positions of the packed beds during discharging pro-
change coefficient, α.
cesses corresponding to A and B, respectively, at H = 1, 2, 3, 4, 5 m.
2.4. Grid independency Velocity of the HTF equals to 0.008 m/s. Inlet temperature of HTF is set
at 250∘C. In both cases of A and B, porosity of the bed is 0.4. Due to the
The grid independency test was carried out for the tank filled with fact that each PCM is located at a given position of the bed, it is clear
same kind of PCM (NaNO2) with the porosity coefficient of 0.4 and that the phase change process of each of them occurs at specified po-
diameter of PCM capsules are 25 mm. The grid is not uniform and at sition and temperature range. Temperature distributions of both layouts
sensitive regions (near the boundaries of capsules and near the walls) are presented in Figs. 5 and 6 and the trends are consistent with results
the mesh grids are smaller so the precision would be higher. Three of [34] and [21].
various grid sizes were considered: 6077360 (fine), 8354208 (extra As it can be inferred from Figs. 5 and 6, discharging process in the
fine) and 10694371 (extremely fine). Simulations were performed for case A (ascending order) is delayed by 320 s in compared with case B
these grid sizes and found that the extra fine (8354208) and extremely (descending order). It is demonstrated in the Fig. 7 that the case B
fine (10694371) mesh cases were producing the same results. Hence, provides a rather constant heat transfer flow in the HTF flow direction
the extra fine mesh was chosen to carry out the heat transfer analysis of which results in faster rate of phase change and completion of dis-
the TES system. Although, as mentioned in Section 2.3, for the mon- charging process.
itored heights the extremely fine mesh was used to utterly enhance the Although this may sound desirable, but there is a down side to faster
precision of the calculations. Fig. 2 is a zoomed-in presentation of the discharging process and that is lower utilization time. Generally, in
mesh grids which have been used here. concentrated solar power plants, the HTF leaving the tank during

Table 2
2.5. CFD model validation
Thermophysical properties of used PCMs.

In order to verify the present simulation, differences of predicted Parameters NaNO2 [21] KNO3 [32] KOH [32].
numerical results and previous experimental results were measured.
Melting temperature ( C)
∘
277.32–303.65 333–336 378–380
Experimental results of Barrientos et al. [33] were used to validate the Latent heat ± 7.5% (Kj/Kg) 199.6 266 149.7
present model. They implemented a packed bed LHS to conduct their Specific heat capacity (Kj/ Solid 1.385 1.439 1.47
experiments. The storage bed was filled with a composite PCM having Kg.K) Liquid 1.531 1.48 1.47
fine particles which make a porosity of 0.486. In this case, mean dia- Density (Kg/m3) Solid 2097 1900 2044
Liquid 1846 1890 2044
meter of PCM spherical capsules were assumed 1.64 mm and air was
Heat conductivity (W/m.K) Solid 0.562 0.5 0.5
used as HTF. Parts (a) and (b) at Fig. 3 compares the numerical and Liquid 0.483 0.5 0.5
experimental temperature distribution of capsules at H = 25 mm and

4
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Fig. 2. Mesh grids used for simulation.

discharge process is fed into the power block for superheated steam As it is clear from Fig. 8, the time periods during which HTF leaves
generation. Since the power block cycle process depends on the HTF the tank with temperature higher than 277∘C, in all of the velocities that
temperature, it requires the termination of discharge process when the have been studied, in case B are approximately 15% higher than case A.
HTF exit temperature reaches minimum cut-off temperature, Td, [35].
The minimum cut-off temperature is determined by the application, in 3.2. Influence of inlet velocity
this investigation Td is assumed to be 277 ∘C. Thus, faster rate of dis-
charging leads to lower performance time for a CSP unit which is un- Effect of the HTF inlet velocity on the thermal performance of the
desirable. packed bed is investigated in this section. For the packed bed of case B,

Fig. 3. Comparison between the present simulation results and experimental results of Barrientos et.al [26].

5
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

time, only half of the tank has been presented. Three different inlet
velocities have been studied and as it can be inferred increasing the
inlet velocity increases the rate of phase change in the PCM spheres and
it results in faster discharging of the packed bed which agrees well with
the experimental and numerical results acquired by Invalid source
specified [33] and [26], respectively.

3.3. Influence of porosity of the packed bed

In this section the effect of different porosities on the discharge

process have been investigated. Three packed bed with identical mea-
surements and conditions have been filled with spheres with porosity
coefficients of 0.4, 0.5 and 0.6.
Fig. 11 shows that decreasing the porosity increases duration of
discharge of HTF at desired temperature. Then by increasing the por-
osity, the number of the spheres containing PCM will decrease and as a
result the amount of energy stored in the bed will also be decreased
which would require less time to complete the process. The results
obtained here are consistent with the results reported in [16] regarding
the influence of porosity coefficient on packed bed discharging.

3.4. Different layouts of the PCMs in the packed bed

As it was mentioned before, the aim of studying packed beds during

the discharge process is to increase duration of recovering energy with
Fig. 4. Two different kinds of placement of three PCMs in the packed bed. desired temperature. In order to achieve this important goal, new
methods of placement of capsules are presented in this paper. Results
achieved from these configurations in beds are favorable. In order to
we have investigated five different velocities and reported the results in
better compare the thermal performance of each layout, total energy
Fig. 9 in which Uf=0.008 m/s. It is demonstrated that by increasing the
utilization, Δtutl, is used. Order of placements of PCMs, as shown in
inlet velocity, the time required for completion of discharge process
Fig. 12, are of case A. Properties of each layout are demonstrated in
decreases and while this can be a good effect in the case of the appli-
Table 3.
cations that fast discharge process is desired, but the total time of
having HTF outlet with temperature higher than 277∘C, will drastically Duration of receiving energy with desired temperature
tutl (%) = × 100
decrease, making it not suitable for applications that require desired Total discharge time of the packed bed
HTF temperature.
(20)
Fig. 10 depicts phase change contours that have been predicted by
the used software. The first layout has the layers of same height and porosity which are
and 0.4, respectively. This layout is the one which was being studied
1
Due to symmetry of the geometry of the tank, to save calculation 3
H
in Section 3.1 and its behavior has been demonstrated in Fig. 13. It will

Fig. 5. Temperature distributions of HTF and capsules in different axial positions of the packed beds during discharging cycles corresponding to A bed.

6
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Fig. 6. Temperature distributions of HTF and capsules in different axial positions of the packed beds during discharging cycles corresponding to B bed.

Fig. 7. Total convection against time in packed beds of layout A and layout B.

7
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Fig. 8. Comparison of time of exiting HTF with a temperature higher than 277∘C for beds of A and B against different velocities.

Fig. 9. Influence of different HTF inlet velocity on duration of discharge process bed A.

be the layout that other ones will be compared to. In layout 1, HTF reason, configuration of case A has been chosen to utterly investigate
flows through three stages of PCM spheres and after exchanging utilization of this case. The total utilization factor for this layout is
thermal energy in each layer, leaves the tank until the process is calculated below so the other layouts’ utilization factor could be com-
complete and the packed bed is discharged. In this section, duration of pared to it:
HTF outlet with desirable temperature is important and because of this

8
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Fig. 10. Comparison of phase change contours of half of the tank with three
different velocities at 800 s.

2300(s)
tutl,1 (%) = × 100 = 46%
5000(s) (21)

In layout 2 the height of each layer is equal to 3 H as before, but the

1

porosities of layers are set in an ascending order of 0.4, 0.5 and 0.6 in
flow direction, respectively. It is shown that although the completion of
discharging process is delayed in compared with layout 1, but the
duration that HTF leaves the tank with desirable temperature has in-
creased. Fig. 12. Different packed bed layouts.
In this layout, after HTF leaves the first layer, it experiences a layer
with higher porosity and less energy; so it leaves the second layer with
lower temperature than same height at layout 1 tank and after entering
the third layer, HTF experiences yet another layer with higher porosity.

Fig. 11. Comparison of discharging completion time of different porosities.

9
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

Table 3 Table 4
Properties of each layout in Fig. 12. Summary of performance of the configurations.
Layouts Layers Layer 1 [KOH] Layer 2 [KNO3] Layer 3 [NaNO2] Layouts Δt Δt(S) Δt(S) Δtutl(%)

Layout 1 Height 1
H
1
H
1
H 1 2300 5000 46%
3 3 3
2 3400 68%
Porosity 0.4 0.4 0.4
3 2810 56%
Layout 2 Height 1 1 1
3
H
3
H
3
H 4 3760 76.2%
Porosity 0.4 0.5 0.6
Layout 3 Height 5
H
3
H
2
H
faster discharge process, but its Δtutl is 11.8% less than layout 2.
10 10 10
Porosity 0.4 0.4 0.4
Layout 4 Height 5 3 2
H H H 2810(s)
10 10 10
tutl,3 (%) = × 100 = 56.2%
Porosity 0.4 0.5 0.6 5000(s) (23)
The forth layout has different heights of H , 10 H
and and
5 3 2
10 10
H
These increases in porosity of each layer in the flow direction causes porosities of 0.4, 0.5 and 0.6 in the flow direction, respectively. This
the HTF to only receive thermal energy needed to raise its temperature layout is a combination of second and third layout. The HTF after
above the desired temperature and to some extent avoid unnecessary leaving the first layout is thermally worn out, even more so in com-
temperature rise. This layout offers 22% better performance in utilizing pared with previous layouts, and then enters the other layers with
thermal energy stored (Δtutl) in comparison with layout 1. higher temperature than what it is supposed to and as a result, it is
unable to optimally experience heat transfer process with PCM spheres.
3400(s) This layout has 29.2% better thermal utilization than first layout and by
tutl,2 (%) = × 100 = 68%
5000(s) (22) far the most optimum layout in compared with other three:
3760(s)
Fig. 13 reports the thermal performance of layout 3 where the tutl,4 (%) = × 100 = 75.2%
5000(s) (24)
porosities of each layer is equal to 0.4 and each layer's height decreases
in the flow direction. After HTF goes through the first layer which oc- A summary of the results of the four configurations is presented in
cupies half of the tank, it leaves the first layer with higher temperature Table 4. According to conditions of using energy storage beds and ex-
than what we remember in the first layout. Then in the second layer, pectations from the beds, one can change the capsule type, porosity and
the HTF that has had its thermal capacity worn out is not able to ex- the height filled with each capsule, to reach the best efficiency.
perience optimal heat transfer and in the third layer the same situation Presenting these configurations in Fig. 13 shows that only by
also appears. An Increase in layer height is similar to a decrease in making little changes in the packed bed configuration and without any
porosity of layer and both mean more energy; only that in the case of need to change the PCMs, it is possible to help enhance the efficiency of
layer porosity reduction the amount of energy raise is more. This layout storing process. These investigations can help better and more effi-
has 10.2% better thermal energy utilization than layout 1, but if it is ciently utilize PCM packed beds and reach better performances in PCM
compared to the second layout, it can be seen that although it has a industry in future.

Fig. 13. Evaluating different configurations against each other.

10
F. Mohammadnejad and S. Hossainpour Journal of Energy Storage 28 (2020) 101209

4. Conclusion [8] H. Singh, R.P. Saini, J.S. Saini, A review on packed bed solar energy storage sys-
tems, Renew. Sustain. Energy Rev. 14 (2010) 1059–1069.
[9] Cabeza Luisa A., Steven Tay Nguan H., High Temperature Thermal Storage Systems
A CFD model was developed to investigate the discharge process of Using Phase Change Materials, Elsevier, 2018.
a packed bed of latent heat storage system. The simulation is developed [10] P.A. Galione, C.D. Pérez-Segarra, I. Rodríguez, A. Oliva a, J. Rigola, Multi-layered
by COMSOL multi-physics 5.2 platform for applications that use high solid-PCM thermocline thermal storage concept for CSP plants. Numerical analysis
and perspectives, Appl. Energy 142 (2015) 337–351.
temperature phase change materials. The effect of placement of PCMs [11] K.A.R. Ismail, R. Stuginsky Jr., A parametric study on possible fixed bed models for
inside the tank according to their phase change temperatures in order of pcm and sensible heat storage, Appl. Therm. Eng. 19 (1999) 757–788.
ascending and descending in flow direction is examined. The influence [12] K.A.R. Ismail, J.R. Henrı́quez, Numerical and experimental study of spherical
capsules packed bed latent heat storage system, Appl. Therm. Eng. 22 (2002)
of inlet velocity of HTF on completion of discharge process is studied
1705–1716.
and it is found that by increasing inlet velocity of HTF the packed bed [13] L. Geissbühler, M. Kolman, G. Zanganeh, A. Haselbacher, A. Steinfeld, Analysis of
depletes more rapidly. The influence of three porosity coefficients in industrial-scale high-temperature combined sensible/latent thermal energy storage,
Appl. Therm. Eng. 101 (2016) 657–668.
packed bed is also studied and as expected it is observed that increasing
[14] A. Felix Regin, S.C. Solanki, J.S. Saini, An analysis of a packed bed latent heat
the porosity of the packed bed decreases the discharge completion time. thermal energy storage system using PCM capsules: numerical investigation,
Furthermore, this paper tried to improve the thermal performance of Renew. Energy 34 (2009) 1765–1773.
packed beds by finding an optimum configuration of layers inside the [15] E. Oró, J. Chiu, V. Martin, F. Luisa, Cabeza, "Comparative study of different nu-
merical models of packed bed thermal energy storage systems, Appl. Therm. Eng.
tank of PCMs. The comparison of the results from five monitored 50 (2013) 384–392.
heights suggests that the configuration that has the PCMs in a des- [16] S. Wu, G. Fang, Dynamic performances of solar heat storage system with packed bed
cending order completes discharge process faster. However, the con- using myristic acid as phase change material, Energy Build. 43 (2011) 1091–1096.
[17] M. Wu, C. Xu, Y.-L. He, Dynamic thermal performance analysis of a molten-salt
figuration with ascending order of PCMs delivers HTF with desired packed-bed thermal energy storage system using PCM capsules, Appl. Energy 121
temperature for a longer period of time. The results show that higher (2014) 184–195.
porosity coefficient and higher HTF inlet velocity lead to faster dis- [18] S. Wu, G. Fang, X. Liu, Dynamic discharging characteristics simulation on solar heat
storage system with spherical capsules using paraffin as heat storage material,
charge process and lesser time of HTF leaving the tank with desired Renew. Energy 36 (2011) 1190–1195.
temperature. The results indicate that by applying a few minor changes [19] S. Karthikeyan, G. Ravikumar Solomon, V. Kumaresan, R. Velraj, Parametric studies
in the order of PCMs, porosity of each layer and height of each layer on packed bed storage unit filled with PCM encapsulated spherical containers for
low temperature solar air heating applications, Energy Convers. Manage. 78 (2014)
inside the tank of packed bed, one can obtain a thermal performance
74–80.
that can better serve desired cause. This demonstration shows that in [20] G. Alvaro de, F. Luisa, Cabeza, "Numerical simulation of a PCM packed bed system:
each layer, there is no need to have the HTF completely drain the en- a review, Renew. Sustain. Energy Rev. 69 (2017) 1055–1063.
[21] H. Peng, H. Dong, X. Ling, Thermal investigation of PCM-based high temperature
ergy of spheres of PCM, but it is only needed to raise the temperature of
thermal energy storage in packed bed, Energy Convers. Manage. 18 (2014)
the HTF to the point that is needed. By doing so, HTF with desired 420–427.
temperature can be obtained for longer time. In this study, it became [22] Z. Liao, G. Zhao, C. Xu, Cenyu Yang ⁎, Y. Jin, X. Ju, X. Du, Efficiency analyses of
possible to optimize the thermal performance of the packed bed by high temperature thermal energy storage systems of rocks only and rock-PCM
capsule combination, Sol. Energy 162 (2018) 153–164.
29.2%. The authors hope to continue this work by providing a three- [23] B. Cárdenas, N. León, High temperature latent heat thermal energy storage: phase
dimensional simulation of this subject with more detailed physical change materials, design considerations and performance enhancement techniques,
modeling. Renew. Sustain. Energy Rev. 27 (2013) 724–737.
[24] S. Jegadheeswaran, S.D. Pohekar, Performance enhancement in latent heat thermal
storage system: a review, Renew. Sustain. Energy Rev. 13 (2009) 2225–2244.
Declaration of Competing Interest [25] T. Watanabe, H. Kikuchi, A. Kanzawa, Enhancement of charging and discharging
rates in a latent heat storage system by use of PCM with different melting tem-
peratures, Heat Recovery Syst. CHP 13 (1) (1993) 57–66.
On behalf of all authors, the corresponding author states that there [26] T. Watanabe, A. Kanzawa, Second law optimization of a latent heat storage system
is no conflict of interest. with PCMS having different melting points, Heat Recovery Syst. CHP 15 (7) (1995)
641–653.
[27] L. Yang, X. Zhang, Performance of a new packed bed using stratified phase change
Acknowledgements
capsules, Int. J. Low-Carbon Tech. 7 (2012) 208–214.
[28] L. Yang, X. Zhang, G. Xu, Thermal performance of a solar storage packed bed using
The authors gratefully acknowledge the Reviewers and editors for spherical capsules filled with PCM having different melting points, Energy Build. 68
(2014) 639–646.
the valuable comments and suggestions.
[29] G. Peiró, J. Gasia, L. Miró, C. Luisa F., Experimental evaluation at pilot plant scale of
multiple PCMs (cascaded) vs. single PCM configuration for thermal energy storage,
References Renew. Energy 83 (2015) 729–736.
[30] N. L. N., Assessment of latent heat thermal storage systems operating with multiple
phase change materials, J. Energy Storage 23 (2019) 442–455.
[1] M. Kenisarin, K. Mahkamov, Solar energy storage using phase change materials, [31] A.B. Zavoico, Solar Power Tower Design Basis Document. Sandia National
Renew. Sustain. Energy Rev. 11 (2007) 1913–1965. Laboratories, Albuquerque, New Mexico, 2001.
[2] I. Dincer, C. Acar, A review on clean energy solutions for better sustainability, Int. J. [32] B. Zalba, J.M. Marin, L.F. Cabeza, H. Mehling, H. Mehling, Review on thermal
Energy Res. 39 (5) (2015) 585–606. energy storage with phase change: materials, heat transfer analysis and applica-
[3] T. Kousksou, P. Bruel, A. Jamil, T. El Rhafiki, Y. Zeraouli, Energy storage: appli- tions, Appl. Therm. Eng. 23 (2003) 251–283.
cations and challenges, Sol. Energy Mater. Sol. Cells 120 (2014) 59–80 VolsPart A. [33] M. Izquierdo-Barrientos, C. Sobrino, J. Almendros-banez, Thermal energy storage in
[4] T.U. Daim, X. Li, J. Kim, S. Simms, Evaluation of energy storage technologies for a fluidized bed of PCM, Chem. Eng. J. 230 (2013) 573–583.
integration with renewable electricity: quantifying expert opinions, Environ. Innov. [34] S. Bellan, J. Gozalez-Aguitar, M. Romero, M. Rahman, D. Goswami, E. Stefanakos,
Soc. Trans. 3 (2012) 29–49. D. Couling, Numerical analysis of charging and discharging performance of a
[5] S. Hasnain, Review on sustainable thermal energy storage technologies, part I: heat thermal energy storage system with encapsulated phase change material, Appl.
storage materials and techniques, Energy Convers. Manage. 39 (11) (1998) Therm. Eng. 71 (2014) 481–500.
1127–1138. [35] K. Nithyanandam, R. Pitchumani, A. Mathur, Analysis of latent thermocline storage
[6] K. Pielichowska, K. Pielichowski, Phase change materials for thermal energy sto- system with encapsulated phase change materials for concentrating solar power,
rage, Prog. Mater. Sci. 65 (2014) 67–123. Appl. Energy 113 (2014) 1446–1460.
[7] H. Mehling, L.F. Cabeza, Heat and Cold Storage with PCM, Springer, 2008.

11