Energy 114 (2016) 10e23

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Novel process for performance enhancement of a solar continuous

adsorption cooling system
Abdellah El Fadar
Laboratory of Innovating Technologies, National School of Applied Sciences, Tangier, Abdelmalek Essaadi University, Morocco

a r t i c l e i n f o a b s t r a c t

Article history: Adsorption cooling is regarded as one of promising clean technologies since it constitutes a rational way
Received 21 April 2016 for the use of low-grade thermal energy (industrial waste, excess production, etc.) and renewable energy
Received in revised form sources. The current paper provides fresh ideas on the relevant issues associated with the intermittence
9 July 2016
and low performance, which are considered among the main drawbacks of solar-driven adsorption
Accepted 27 July 2016
cooling systems. Its primary aim is to design and analyze a solar continuous adsorption cooling system
whose design consists mainly of combining a two-tank thermal energy storage system with two
adsorbent beds. The operating process is based on storage of excess thermal energy, supplied by means
Keywords:
Thermal energy storage
of a solar parabolic trough concentrator, and on its subsequent recovery for producing an additional
Solar parabolic trough collector amount of cold.
Cooling The simulation results indicate that, under the system conditions (design, working and weather), the
Adsorption solar coefficient of performance and daily cooling production are considerably improved when a latent
Continuous cycle heat storage unit is integrated with the system. The study reveals also that, thanks to this novel process,
Modeling the number of refrigeration cycles achieved per day could be significantly increased, which means that
the intermittence disadvantage could widely be overcome.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction environmental problems since the adsorption cooling systems use

Current trends in energy supply and use are patently unsus-
tainable e economically, environmentally and socially [1]. Given
this harsh reality, the policymakers and scientific community are
increasingly confronted with the complex challenges of sustainable
development, energy security and climate change.
Heating, ventilation and air conditioning (HVAC) are among the
energy uses that consume a significant fraction of energy and
release large amounts of greenhouse gas. Indeed, the IEA (Inter-
national Energy Agency) reported, in 2015, that: “today, heating
and cooling in buildings and industry accounts for approximately
40% of final energy consumption. With 70% of heating and cooling
demand relying on fossil energy sources, these end uses are esti-
mated to have been responsible for 30% of global carbon dioxide
(CO2) emissions in 2012” [2].
In field of refrigeration and air-conditioning, the adsorption
cooling technology is more and more receiving attention and is
regarded as a green technology to lessen the above energetic and
E-mail addresses: elfadar@ensat.ac.ma, aelfadar@yahoo.fr.
environment friendly refrigerants (water, methanol, ammonia,
etc.), they could operate directly with a primary source of energy,
such as solar and geothermal energies, and they could also use
industrial waste heat [3]. Furthermore, they could operate at a
driving temperature (52.5e82
C) lower than that of absorption
systems [4,5], and they are not noisy due to their ability to work
without moving parts. In this context, the solar adsorption cooling
technology appears to be promising due mainly to the cleanness
and abundance of the solar energy, and because of the close coin-
cidence between the availability of solar radiations and the peak of
cooling demands. So, applications of this technology are especially
interesting in sunny (i) non-electrified regions, where the preser-
vation of foods and pharmaceutical/medical products is vital, and
(ii) electrified areas in order to save energy and mitigate carbon
emissions.
The adsorption refrigeration processes are carried out using a
certain number of working pairs (adsorbent/refrigerant). The
common physical adsorption pairs used in this field include silica
gel/water, zeolite/water, activated carbon (AC)/methanol, AC/
ammonia, AC/ethanol, etc. It is to mention here the shortcomings of
methanol and ammonia, namely flammability and toxicity,

http://dx.doi.org/10.1016/j.energy.2016.07.149
0360-5442/© 2016 Elsevier Ltd. All rights reserved.
 A. El Fadar / Energy 114 (2016) 10e23 11

Nomenclature le equivalent thermal conductivity of adsorbent

(W m1 K1)
Ac collector aperture area (m2) r density (kg m3)
AHWST area of hot water storage tank (m2)
cp specific heat capacity (J kg1 K1) Subscripts
ha latent heat of adsorption (J kg1) 0 without latent heat storage
hr hour (s) 1 with latent heat storage
Ib direct solar irradiance on collector aperture (W m2) a adsorbate, adsorbed
Ibn direct normal solar irradiance (W m2) ads adsorption
Lc collector length (m) amb ambient
Lr reactor (adsorber) length (m) con condensation, condenser
Lvap vaporization latent heat of ammonia (J kg1) ev evaporation, evaporator
m mass (kg) f fluid (water) in hot water storage tank
ma adsorbed mass on a layer of adsorbent (kg) fin final
m_ mass flow rate (kg s1) g gas
N number of cooling cycles (dimensionless) i initial
P pressure (Pa) in inlet
Qc daily cooling production (J) l liquid (ammonia)
qc cooling production during one cooling cycle (J) m melting
r radial coordinate (m) max maximum
T temperature (K) met metal
Tg1 temperature at the start of desorption process (K) me metallic tube in adsorbent
Tg2 temperature at the end of desorption process (K) out outlet
Tsuit suitable driving source temperature (K) rec recovered
t time (s) sat saturation
t1 time at the beginning of cooling production process sav saved
without latent heat storage (s) sr sunrise
t2 time at the end of cooling production process without ss sunset
latent heat storage (s) st0 storage in hot water storage tank without any heat
t3 time at the end of cooling production process with removal
latent heat storage (s) st1 storage in hot water storage tank with heat removal
tcycl cycle time (s)
U heat loss coefficient from hot water storage tank Abbreviations
(W m2 K1) AC activated carbon
W width of collector aperture (m) CWST cold water storage tank
DSCP daily specific cooling production (J kg1)
Greek symbols HTF heat transfer fluid (oil)
Dx adsorption capacity difference between adsorption HWST hot water storage tank
and desorption phases (kg kg1) LHS latent heat storage
ε porosity of adsorbent bed (dimensionless) PCM phase change material
h latent heat storage efficiency (dimensionless) PTC parabolic trough collector
hc collector efficiency (dimensionless) SCOP solar coefficient of performance
q volume fraction of the adsorbed phase (dimensionless) SHS sensible heat storage
qi incidence angle (
) TES thermal energy storage

respectively. An extensive review about the selection criteria of
working pairs could be found in Ref. [6]. It is worthy to note that
each one of these pairs requires a driving source temperature that
should be well chosen to generate a large amount of refrigerant.
The choice of regeneration temperature will be discussed below in
Section 2.
Although the adsorption chillers, which first appeared on the
market in 1986 by the Nishiyodo Kuchouki, Co. Ltd [7], offer the
above advantages, their expansion in the market remains still
limited because of some technical limitations such as intermittence
in operation when driven by solar energy, bulkiness, low perfor-
mance and low specific cooling power due to weak heat and mass
transfer in adsorbents. To improve the performance of the
adsorption bed, efforts should be focused on reducing the inter-
particle thermal resistances as well as the interior mass transfer
resistances of the particle [8,9]. Comparatively, the inter-particle
mass transfer resistance is less important [9].
On the other hand, the parabolic trough collectors (PTCs) have
been used in numerous applications, such as electricity generation,
desalination, heating [10], whereas very few works have been
devoted to cooling purposes in spite of their benefits, namely they
can achieve higher tracking accuracy than dish-engine collectors
[11], they are lighter than flat plate and evacuated tube collectors;
moreover, their high efficiency may offer the possibilities to pro-
duce high amounts of thermal energy that could be stored and
subsequently used in an effective way depending on the applica-
tion. These advantages could be a satisfactory solution to overcome
the low performance, bulkiness and intermittence drawbacks of
the solar adsorption cooling systems.
In fact, continuous solar adsorption cooling systems, based upon
two or more adsorbers, are more interesting due to the timely
coincidence that offer between needs and production of cold, and
12 A. El Fadar / Energy 114 (2016) 10e23
also due to their higher performance over intermittent ones which
produce cooling only during nighttime. In an attempt to overcome
intermittence drawback and hence reduce the mismatch between
supply and demand of cold, a limited number of research works
have been devoted to developing continuous/quasi-continuous
solar adsorption refrigeration cycles.
Some remarkable works have been reported in literature: Zhang
and Wang [12] designed a solar continuous adsorption refrigeration
and heating hybrid system. In order to produce cooling continuously,
the upper bed of the two-bed system is heated up by solar radiations,
while the lower bed is cooled down from a cold water storage tank.
When the two beds are in a state of saturated desorption/adsorption
respectively, their positions are shifted through a rotation of 180
by
an axis. The system could produce 30 kg of hot water at a temper-
ature of 47.8
C, while the specific cooling power was found to be
17.6 W/kg with a solar cooling COP of 0.18.
A hybrid system for refrigeration and water heating was also
investigated by Alghoul et al. [13]. In order to produce ice contin-
uously, the researchers used two water storage tanks each of them
contains an adsorber and a condenser. Adsorber 1 in tank 1 is
connected to condenser 2 in tank 2 and vice versa. The water in
storage tanks is heated using evacuated tube collectors. During
daytime, desorption occurs from adsorber 1 to the receiver through
condenser 2 while adsorption process takes place from the receiver
to the adsorber bed 2 through condenser 1. In the evening, the roles
of the two adsorbers are inverted. They reported that the system is
capable to provide a quantity of hot water for domestic use of
116 kg/day and an amount of ice of 12 kg/day.
El Fadar et al. [14] designed and investigated the performance of
a continuous adsorption cooling system driven by parabolic trough
solar collector (PTC), whose process was based on sensible heat
storage (hot water). For producing cold continuously, the two ad-
sorbers were heated up and cooled down alternatively from a hot
water storage tank and a cold water storage tank, respectively. The
authors reported that the system could achieve a specific cooling
power of about 104 W/kg, a refrigeration cycle COP of 0.43 and a
solar COP of 0.18.
Zhai and Wang [15] have experimentally tested a solar-powered
adsorption cooling system which can be switched between a sys-
tem with heat storage and a system without heat storage. Results
indicated that, the system with heat storage operated stably due to
the regulating effect by the heat storage water tank compared to
the system without heat storage. However, under similar condi-
tions, the system without heat storage is capable of achieving a
similar cooling effect to the system with heat storage although it
has an obvious varying character. The system without heat storage
is more suitable for areas with abundant solar energy resources.
Li et al. [16] have experimentally tested a solar adsorption ice-
maker driven by PTC and integrated with sensible thermal storage
tank. The results showed that the highest COP reached 0.15 while
the solar coefficient of performance could be 0.08 and the ice
making capacity was 50 kg per day with 20 m2 of PTC and 30 kg of
compound adsorbent (calcium chloride/activated carbon) when
the desorption temperature and the direct normal solar radiation
were 105
C and 3 kW h/day m2, respectively.
Abu-Hamdeh et al. [17] investigated an improved prototype of
an adsorption cooling system. The system used olive waste/meth-
anol as working pair. A storage tank was used to store hot water
produced by a PTC. Hot water passes then through the bed to heat
up the adsorber in order to generate desorption of methanol. The
results obtained showed that the optimal adsorbent mass varied
between 30 and 40 kg and the optimum tank volume was found to
be between 0.2 and 0.3 m3, while the optimum collector area varied
between 3.5 and 5 m2. The solar COP of the system varied from 0.18
to 0.2.
Later, El-Sharkawy et al. [18] have proposed two system con-
figurations: in the first one, adsorption chiller is directly connected
with a solar compound parabolic collector (CPC) whilst the second
one includes a hot water buffer storage that is installed between
adsorption chiller and solar collector. According to the authors, the
results of this study show that the second configuration allows
reducing fluctuations of hot water temperature and producing
higher cooling capacity at the beginning and the end of the day
time.
Recently, in 2015, Deshmukh et al. [19] presented a solar pow-
ered continuous adsorption cooling system with three adsorbent
beds, in which cold energy is stored in the form of liquid refrigerant
at ambient temperature. They concluded that the system is capable
to provide cooling capacity of about 0.8 kW for 24 h. The specific
cooling effect and average cycle COP are found to be 337.5 kJ/kg and
0.63 respectively.
From this literature survey, the most reports in this area
revealed that it is feasible to achieve an uninterrupted/quasi-
continuous operation of solar adsorption refrigeration system if
solar energy could be stored in the form of hot water i.e. sensible
heat storage (SHS). Nevertheless, latent heat storage (LHS) using
phase change materials (PCMs) offers some advantages over sen-
sible heat storage, such as higher heat storage densities over a
narrow operating temperature range between storing and releasing
heat [20,21]. In addition, LHS is a rational approach for thermal
energy management, since the heat storage could be achieved from
renewable energy sources, industrial waste heat or excess pro-
duction of thermal energy.
The use of PCMs for thermal energy storage (TES) is limited
however by the high thermal resistance due to their low thermal
conductivity, leading to low charging and discharging rates [22]:
during discharge process, a layer of solid PCM forms on the heat
exchange surface, progressively reducing the heat transfer rate
[23]. This weakness can be overcome by some techniques, such as
finned tubes of different configurations, bubble agitation, shell and
tube (multi-tubes), micro-encapsulating the PCM, insertion of
metal matrix into the PCM and the use of heat pipes [24].
In this respect, various groups of PCMs with large range of
melting temperatures could be used in diverse applications such as
thermal comfort building, solar heating system, thermal protection,
air-conditioning, transportation, electronic devices, etc. [25]. The
ideal PCM should meet a number of criteria related to the required
thermophysical (suitable phase-change temperature in the desired
operating temperature range, small volume change on phase-
change, high latent heat of transition per unit mass, high thermal
conductivity and good heat transfer, etc.), kinetic (high nucleation
rate and little or no supercooling of the liquid phase, high rate of
crystallization), chemical (complete reversible melt/freeze cycles,
stability and no degradation, no corrosiveness, nontoxic, non-
flammable, etc.), economic (abundant, inexpensive), and environ-
mental (low environmental impact and non-polluting, low
embodied energy, recycling potential, etc.) properties [26].
On the basis of the previous analysis, the current work focuses
on the following novelties: (i) introduction of a new process in
adsorption cooling based on both latent and sensible heat storages
of solar energy, (ii) use of the excess heat for producing an addi-
tional cold, contrary to the existing works based on heat storage
where the surplus heat is mostly utilized for domestic water
heating, this issue is relevant especially in hot areas where cooling
needs are much higher than heating requirements, and (iii) regu-
lation and control of regeneration temperature for desorbing a
large amount of adsorbate and for energy saving. This study pro-
vides useful ideas to researchers, engineers and manufacturers who
are interested in promoting this green technology (adsorption
cooling). Through the modeling and analysis below, we aim to put
 A. El Fadar / Energy 114 (2016) 10e23 13

in evidence that the proposed process could achieve significant
improvements of performance.
2. System description and operating principle
The proposed system is schematized in Fig. 1. It consists
essentially of:
- two adsorbers (1,2) containing the adsorbent/adsorbate (acti-
vated carbon/ammonia) pair;
- a condenser (3);
- an evaporator (4);
- a liquid refrigerant tank (5);
- a hot water storage tank (6);
- a cold water storage tank (7);
- a parabolic trough concentrator (8);
- a latent heat storage unit containing a phase change material
(9);
- a differential temperature regulator (10);
- three circulation pumps (11,12,13); and
- valves (14,15).
Thermal energy collected by the parabolic trough concentrator
(PTC) is transferred to the thermal oil as heat transfer fluid (HTF)
flowing through the absorber tube (16), this one is positioned at the
focal line of PTC and surrounded by a glass envelope in order to
reduce heat losses to the ambient environment. The heat provided
by the PTC to the HTF is stored in hot water storage tank (HWST) via
a heat exchanger. The hot water is then used to generate desorption
of refrigerant (ammonia) from adsorbents, while the cold water
stored in cold water storage tank (CWST) is utilized to generate
adsorption in adsorbents. The two cylindrical adsorbers (1,2),
which are assumed to be perfectly insulated at their outer areas, are
alternatively heated up/cooled down by means of the hot/cold
water, circulating through a stainless steel tube in each adsorber,
from the HWST/CWST i.e. when an adsorber (1 or 2) is heated up to
desorb the refrigerant towards the condenser, the other adsorber (2
or 1) is cooled down to adsorb the refrigerant coming from the
evaporator.
In order to develop this process based only on sensible heat
storage, which was proposed in our previous work [14], and
improve the system performance, we suggest in the current study
the integration of a latent heat storage unit containing phase
change material (PCM) into the system. A control and command
system composed mainly of a temperature regulator (10) and a
three-way solenoid valve (15) allowing the HTF crossing the
absorber (16) to flow alternately to (i) the HWST (6) or to (ii) the
PCM unit (9) as follows:
i) the HTF flows towards the HWST (6) via a circulation pump (11)
and the solenoid valve (15), i.e. the HTF circulates through loop
17 (primary cycle), if two conditions are simultaneously met: (1)
the temperature difference between the collector outlet and
inside of the HWST (DT ¼ Tout e THWST) exceeds a beforehand
fixed value and (2) the HWST temperature is lower than the
suitable driving source temperature (for example, DT > 2
C and
THWST  90
C); the choice of Tsuit ¼ 90
C will be clarified later in
this section;
ii) or it flows towards the PCM unit (9) via a second circulation
pump (12) and the solenoid valve (15), i.e. the HTF circulates
through loop 18 (charging cycle), otherwise (DT  2
C or
THWST > 90
C) in order to store excess thermal energy in this
unit during the charging process i.e. during the melting phase of
PCM. The pump (12) is started up as long as the collector outlet
temperature is higher than the temperature of the PCM unit
outlet.

15
1
18 10
16
9 14
8
4 5 3
6

17
12
2
11
13
19

1&2: two adsorbers 10: differential temperature regulator Refrigerant path

3: condenser
4: evaporator
5: liquid refrigerant tank
6: hot water storage tank
7: cold water storage tank
8: parabolic trough concentrator
9: latent heat storage unit
11, 12 &13: circulation pumps Hot water path
14 & 15: valves Cold water path
16: absorber (receiver) of concentrator Oil path
17: primary cycle (loop 11–16–15–6–11)
18: charging cycle (loop 12–16–15–9–12)
19: discharging cycle (loop 13–6–9–13)

Fig. 1. Schematic diagram of the solar powered continuous adsorption refrigeration system.
14 A. El Fadar / Energy 114 (2016) 10e23
Fig. 2. Annual global horizontal irradiation map of Morocco [34].
The heat stored in the PCM unit is afterward recovered via a
third circulation pump (13) during the discharging process (during
the solidification phase of the PCM) through loop 19 (discharging
cycle), when the ambient temperature and solar radiations
decrease. The recovered heat will be used for heating up the water
in the HWST and then to heat up the adsorbent beds in order to
generate desorption.
The electrical energy necessary for operating the various
equipments, namely the solenoid valve (15), pumps (11,12,13),
temperature regulator (10) and engine allowing the PTC to track the
sun, could be produced from renewable energy sources (for
example, photovoltaic panels).
It is clear, from this description, that energy saving is accom-
plishable due to the undertaken thermal management, precisely to
the control of temperature in HWST. Indeed, with a suitable driving
source temperature (Tsuit), large amounts of refrigerant is desorbed,
but beyond this temperature (T > Tsuit), a large amount of heat will
be absorbed by the components of adsorbers (adsorbent, metal)
resulting only in rise of their temperatures and a very small amount
of refrigerant will be desorbed, which constitutes a waste of ther-
mal energy. Accordingly, the driving source temperature should be
well chosen depending on the pair used. For example, it is reported
from literature that the desired temperature for AC/ammonia is in
the range between 90 and 100
C [14]; it was also revealed that the
best thermal performances are obtained with a driving tempera-
ture of 100
C, when the monolithic carbon is used as adsorbent
[27].
The highest heating temperature is required for zeolite/water
pair, which is about 250e300
C [28]. From Ref. [29], it was re-
ported that the optimum desorption temperature for zeolite/water
pair is about 200
C whilst it is 120
C for AC/methanol pair. For
silica gel/water, this temperature is roughly between 75
C and
90
C [6]; from Ref. [30], this pair can be regenerated at a relatively
low temperature (below 100
C and typically about 85
C); it is also
pointed out that about 95% of regeneration is obtained at 95
C
[19,31e33]. Therefore, we can conclude that for each pair, corre-
sponds an optimal heat source temperature that should be
controlled for generating a maximum quantity of adsorbate from
adsorbent and for energy saving.
For the AC/ammonia pair used in this work, once THWST exceed
Tsuit ¼ 90
C, the pump (11) will turn off and the pump (12) will
simultaneously turn on in order to store thermal energy in the
latent heat tank. The thermal energy that should be recovered
during the discharge process via the pump (13) allows producing
an additional cold.
The proposed system could be used for diverse cooling pur-
poses, particularly in strongly sunned sites, where needs in cold are
high, such as Morocco. Indeed, according the renewable energy
atlas, developed by the ADEREE (National Agency for the Devel-
opment of Renewable Energy and Energy Efficiency) in partnership
with SAS NOVELTIS [34], Morocco has a big irradiation potential
that exceeds 2000 kWh/m2 per year over nearly half of the coun-
try's surface area (710 850 km2) as shown in Fig. 2. Its daily irra-
diation potential is 5.3 kW h/m2 with annual sunshine durations
ranging from 2700 h in the north to approximately 3500 h in the
south. On a global scale, Morocco occupies the 9th position [35].
3. Modeling and theoretical analysis
3.1. Model assumptions
The main assumptions of the current model are as follows:
 The pressure in adsorbent bed is uniform;
 The adsorbent bed is characterized by an equivalent thermal
conductivity;
 The heat transfer in adsorbent bed is radial;
 The adsorption/desorption phases are isobaric;
 The adsorbent bed properties have a cylindrical symmetry;
 Heat loss from the pipes to the surroundings is negligible;
 A. El Fadar / Energy 114 (2016) 10e23 15

The model equations are detailed as follows:

ZTg2
Q3 ¼ mAC xðT; PÞcp;l dT (6)
3.2. Hot water storage tank Tads

3.2.1. Without any heat removal

ZTg2
The energy balance equation in the hot water storage tank, in
Q4 ¼ mAC hd dx (7)
the case where no part of heat is removed from the HWST, can be
expressed by: Tg1

 dT     wherehd represents latent heat of desorption (in kJ kg1), and x(T,P)

st0 _ p HTF Tst0;in  Tst0;out
mf cp;f þ mmet cp;met ¼ mc designates the adsorbate concentration ratio (in kg/kg of AC) at T
dt
 ðUAÞHWST ðTst0  Tamb Þ (1) and P, which is commonly estimated by the DubinineAstakhov
(DeA) equation [36]:
The terms of this equation represent respectively:    n
Psat
x ¼ W0 rl ðTÞexp  D T ln (8)
- sensible heat of the HWST components (water þ metal); P
- thermal energy absorbed by the HWST from the HTF;
- thermal losses from the HWST to the surrounding. where Wo, rl, D and n are the maximum adsorption capacity, the
density of liquid adsorbate, the coefficient of affinity and a
parameter characterizing the adsorption pair, respectively. The
numerical values of Wo, D and n are furnished in Ref. [14].
On the other hand, the instantaneous collector efficiency is
3.2.2. With heat removal
expressed as:
Heat can be extracted from the HWST for heating up the ad-
sorbers as long as the HWST temperature is sufficient (90
C in this    
_ p
mc TPTC;out  TPTC;in
work) to generate desorption from each adsorbent bed. In this case, hc ¼ HTF
(9)
the energy balance equation in the storage tank is written as
Ib Ac
follows:
where Ac is the collector aperture area:

 dT    
mf cp;f þ mmet cp;met st1 _ p HTF Tst1;in  Tst1;out
¼ mc Ac ¼ WLc (10)
dt
 ðUAÞHWST ðTst1  Tamb Þ and
.
 Qabs tcycl (2) Ib ¼ Ibn $cos qi (11)
In Eq. (2), tcycl denotes the cycle time while Qabs represents the Ib is the component of the normal beam irradiance on the plane of
heat supplied to the two adsorbers during one cycle: the collector aperture (sloped surface). Ibn is the direct normal solar
irradiance (direct irradiance received on a plane normal to the sun),
Qabs ¼ Q1 þ Q2 þ Q3 þ Q4 (3) which is usually known; it could be measured with a pyrheliome-
ter.qi is the incidence angle, which is defined as the angle between
where Q1, Q2 and Q3 are the sensible heats consumed by the
the collector aperture normal and the incident solar beam (central
metallic tube (inserted in adsorbent), adsorbent and refrigerant,
ray) from the sun (solar disk). For a PTC, a tracking collector, the
respectively, while Q4 is the latent heat required to desorb refrig-
concentrator aperture should constantly be normal to the sun, the
erant from adsorbent [14]:
incidence angle could therefore be taken equal to 0. This is espe-
cially true with the two-axis tracking system.
ZTg2
Q1 ¼ mme cp;me dT (4)
3.3. Heat and mass transfer equations in the adsorbent bed
Tads

The model describing the combined heat and mass transfer for a
ZTg2 control volume (a layer with radial coordinate r and thickness dr) in
Q2 ¼ mAC cp;AC dT (5) a cylindrical adsorbent bed has been explained in detail in our
previous study [14]. It is expressed by the following equation:
Tads

" #    
h i vT v2 T 1 vT P v
 1 P vma 1 vma
ð1  εÞrAC cp;AC þ ðε  qÞrg cp;g þ qra cp;a ¼ le þ þ ðε  q Þ r þ þ ha (12)
vt
|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} vr 2 r vr rg vt g
2prLr dr ra vt 2prLr dr vt
ð1Þ
|fflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
ð2Þ ð3Þ ð4Þ ð5Þ
16 A. El Fadar / Energy 114 (2016) 10e23

The terms of this equation represent (1) sensible heat of the

adsorbent and ammonia at gaseous and adsorbed phases, (2) heat Ztd
   
exchanged by conduction (3) elastic energy of ammonia gas, (4) Q¼ _ p HTF THTF;out  THTF;in dt
mc (16)
elastic energy of adsorbed ammonia, and (5) heat of adsorption. 0

where td denotes the discharging time.

3.4. Efficiency of latent heat storage unit
The aim of this paragraph is to assess the amount of heat
extracted from the HWST to be stored in the PCM unit (saved heat)
and the fraction of the saved heat that could be released from the
PCM unit (recovered heat).
3.4.1. Saved heat
As pointed out earlier, thermal energy could be stored in PCM
unit as long as temperature in hot water storage tank exceeds the
suitable driving source temperature (THWST > Tsuit). Thus, the
amount of surplus heat that could be extracted from the HWST to
be stored in the PCM unit can be evaluated by the following
formula:
 
Qsav ¼ mf cp;f Tst1;max  Tsuit
Tst1, max represents the maximum temperature reached by hot
water in the HWST with heat removal process.
Furthermore, the storage capacity of latent heat storage system
with a phase change material is expressed by Refs. [37,38]:
The heat recovered from a thermal energy storage (TES) system
can also be evaluated depending on TES energy efficiency, which
can be defined in various ways, for example [43]:
Energy recovered ðfrom TESÞ
h¼
Energy input ðto TESÞ
More simply, through a comparative study between the effi-
ciencies of three kinds of TES, it has been estimated that phase
change materials can offer storage efficiencies from 75 to 90%
[44,45], see Table 1.
In recent literature, there is a trend of development of PCMs
with higher heat storage capacities. For example, it was designed a
novel type of bifunctional microencapsulated PCM with an excel-
lent latent-heat storage and release performance; the thermal
storage capabilities were estimated to be higher than 98% [46].
(13)
In light of this analysis, we will investigate in Section 5, the
cooling system performance in function to latent heat storage ef-
ficiency values in the range between 0.50 and 0.90. The recovered
heat will be estimated by:
Qrec ¼ hQsav (17)

ZTm ZTfin where h stands for the latent heat storage efficiency.
   
Qst ¼ mcp PCM
dT þ fm mPCM DHm þ mcp PCM
dT (14)
Ti Tm
3.5. Climatic data
where fm designates the melted mass fraction of PCM, while DHm is
the latent heat of melting/solidification per unit mass (in J/kg). Tm is For climatic data to be used in simulation, we consider a typical
the melting temperature that should be in the range of application. summer day under a clear sky, for which the beam solar irradiance
As mentioned before, there are numerous types of PCMs with wide at normal incidence, Ibn, is assumed to be varied as a sinusoidal
range of Tm that could be appropriate for use in process described in function. This function takes into account the sunrise time tsr ¼ 5 h,
the current work, for example [39]: the sunset time tss ¼ 18 h and the maximal direct solar irradiance
Ibn, max ¼ 880 W/m2. These values correspond to direct solar irra-
- Organic phase change materials: paraffins (5.5e75.9
C), non- diance, which was estimated from measured diffuse and global
paraffins (7.8e127.2
C) and fatty acids (16.7e102
C); irradiance for 5 August 2011 [47] corresponding to Tetouan city
- Inorganic phase change materials: metallics (29.8e125
C) and (latitude: 35.57361 N, longitude 5.37528 W) in Northern Morocco.
salt hydrates from 14.0 to 117
C (from 8.1 to 116.7
C according
to Ref. [40]);
- Eutectics (organic and inorganic): (14.7e81.6
C).

3.4.2. Recovered heat

The rate of heat released from latent heat storage system is
given by:
   
_ p HTF THTF;out  THTF;in
q ¼ mc (15)

DTHTF ¼ THTF,out e THTF,in is the difference between inlet and

outlet temperatures of HTF flowing in the PCM unit. The heat
release capacity is given by Refs. [41,42]:

Table 1
Typical parameters of thermal energy storage systems [44,45].

TES system Capacity Efficiency (%) Cost (V/kWh)

(kWh/tonne)

Sensible (hot water) 10e50 50e90 0.1e10

PCM 50e150 75e90 10e50
Chemical reactions 120e250 75e100 8e100
Fig. 3. Climatic data used in simulation.
 A. El Fadar / Energy 114 (2016) 10e23 17

Table 2
Parameters used in simulation.

Symbol Parameter Value Unit

Collector components
Lc collector length 1.50 m
W aperture width of the collector variable (0.6e1.4) m
hc collector efficiency 0.85 e
rmet density of the metal (stainless steel) 7850 kg m3
Hot water storage tank (HWST)
Cp,f specific heat of water 4.18 kJ kg1 K 1
Cp,met specific heat capacity of the metal 0.46 kJ kg1 K1
e tank thickness 0.003 m
UHWST heat loss coefficient from the HWST 2.00 W m2 K1
VHWST volume of the HWST 0.050 m3
AC/ammonia pair
Cp,AC specific heat of adsorbent 0.836 kJ kg1 K 1
Di inner diameter of the metal heat transfer tube 0.034 m
Do inner diameter of adsorbent bed 0.04 m
D1 outer diameter of adsorbent bed 0.10 m
Lr reactor (adsorber) length variable (0.5e6.4) m
Lvap vaporization latent heat of ammonia (J kg1) 1262.40 kJ kg1
ha latent heat of adsorption 1600 kJ kg1
le equivalent thermal conductivity of adsorbent 0.431 W m1 K1
rAC density of adsorbent (AC) 500 kg m3
Operating conditions
Tads adsorption temperature 297.15 K
Tcon condensation temperature 308.15 K
Tev evaporation temperature 273.15 K

8  2 3
>
< I ðtÞ ¼ I pðt  tsr Þ ZTcon
bn; max sin t2½tsr ; tss  6 7
bn
ðtss  tsr Þ (18) qc ¼ mAC Dx4Lvap  cp;l dT 5 (21)
>
:
Ibn ðtÞ ¼ 0 t2½0; tsr ∪½tss ; 24 Tev

The daily ambient temperature depends on the highest (Tmax) where cpl is the specific heat of liquid ammonia; it is given (in J
and lowest (Tmin) values as expressed in the following equation kg1 K1) in the temperature range from e 45 to þ 45
C by the
[48,49]: following equation [51]:
    
Tmax þ Tmin Tmax  Tmin pðt  8Þ 16:842
Tamb ðtÞ ¼ þ sin cp;l ¼ 3:1365  0:00057T þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (22)
2 2 ðtss  tsr Þ 133  T
(19)
Tmax and Tmin are taken to be 42 and 24
C
respectively.
C,

The graphs of climatic data (solar direct irradiance and ambient ➢ The DSCP is defined as the daily cooling production per unit of
temperature) are reported in Fig. 3. The other (design and oper- adsorbent mass:
ating) parameters used in simulation are listed in Table 2.
Qc
DSCP ¼ (23)
mAC
4. Performance indicators and numerical solution
mAC designates the adsorbent mass of the two beds.
4.1. Refrigeration system
4.2. Calculation steps of performance
In this study, the performance of the adsorption cooling system
is assessed through two indicators, namely solar coefficient of
The SCOP and DSCP calculation is detailed below by dis-
performance (SCOP) and daily specific cooling production (DSCP),
tinguishing two cases:
which are defined as follows:

➢ The SCOP is defined as the ratio between the daily cooling 4.2.1. Without latent heat storage unit
production Qc and the useful irradiance falling on the collector
aperture area from sunrise to sunset: - Number of cooling cycles

Qc N qc ðt2  t1 Þ
SCOP ¼ Z sunset
¼Z sunset
(20) N0 ¼ (24)
tcycl
Ac Ib ðtÞdt Ac Ib ðtÞdt
sunrise sunrise
t1 and t2 represent start time and finish time of cooling production
N represents the number of cooling cycles that could be achieved process, respectively i.e. in this case (without LHS), cooling pro-
per day, while qC stands for the cooling production per one cycle. It duction happens only within the time range from t1 to t2, as it will
is expressed by the following formula [50]: be discussed later in Section 5.
18 A. El Fadar / Energy 114 (2016) 10e23

- Daily cooling production

ðt2  t1 Þ
Qc;0 ¼ qc N0 ¼ qc (25)
tcycl

- Daily specific cooling production

Qc;0 ðt  t1 Þ
DSCP0 ¼ ¼ qc 2 (26)
mAC tcycl mAC

- Solar coefficient of performance

Qc;0
SCOP0 ¼ Z sunset
(27)
Ac Ib ðtÞdt
sunrise

4.2.2. With latent heat storage unit

- Number of cooling cycles

Qrec
N1 ¼ N0 þ (28)
Qabs
Fig. 4. Temperature evolution within storage tank with and without heat removal
versus time, Lr ¼ 0.5 m; W ¼ 1.0 m and h ¼ 0.9.

- Daily cooling production

which serve to evaluate the performance parameters (SCOP and
Qc;1 ¼ qc N1 (29)
DSCP) using equations given in Section 4. Validation of the partial
model in adsorbent has been proven in our previous work [14], in
- Daily specific cooling production which the used experimental set-up was described. The flowchart
of the model algorithm, given in Appendix 1 (a and b), indicates the
QC;1 qC N1 main stages of the simulation program.
DSCP1 ¼ ¼ (30)
mAC mAC
5. Results and discussions
- Solar coefficient of performance
5.1. Effect of the LHS unit on the system performance
Qc;1
SCOP1 ¼ Z sunset
(31) The aim of this section is to illustrate the effect of the latent heat
Ac Ib ðtÞdt storage (LHS) unit on the system performance. The results obtained
sunrise
are summarized in Fig. 4 and Table 3. Firstly, for a given aperture
area of collector (Lc  W) ¼ (1.5  1) m2, Fig. 4 depicts the tem-
- Time at the end of cooling production perature evolution of hot water storage tank (THWST) for two cases:
(i) without any heat removal (green line before t1 and red line after
 
Qrec t1) and (ii) with heat removal (green line) for heating up the two
t3 ¼ t2 þ t (32)
Qabs cycl adsorbers and then generating desorption. As can be seen in this
figure, from sunrise time (tsr ¼ 5.00 h), the temperature of storage
tank increases progressively, due to solar irradiation fluctuation.
The maximal temperatures with and without heat removal are
4.3. Numerical solution Tmax1 ¼ 125.34
C and Tmax0 ¼ 156.73
C, which are reached at
tmax1 ¼ 15.85 h and tmax0 ¼ 17.10 h, respectively.
The numerical method used to solve the differential equations Heat removal process is started when the value of suitable
of the model is based on the implicit finite difference scheme. driving source temperature (Tsuit ¼ 90
C) is reached i.e. from time
Iterative techniques are implemented to resolve the nonlinearity of t1 ¼ 11.03 h. This process is continued as long as THWST is higher
Eq. (12). Besides, the implicit Euler method is used to determinate than Tsuit i.e. until t2 ¼ 21.31 h as shown in Fig. 4. Therefore, without
the temperature in hot water storage tank, at each time step: LHS unit, cooling production occurs during 10.28 h (from t1 to t2)
Dt ¼ 30 s, for the two cases: without and with heat removal. The using only the sensible heat storage unit (HWST). However, beyond
times t1 and t2 have been numerically estimated by comparing, at t2 (t > t2), the temperature in HWST becomes lower than Tsuit, so it
each time step, temperature in the HWST with the suitable driving is insufficient to desorb enough amount of refrigerant from
source temperature via numerical tests. adsorbent and hence no significant cooling is produced. In this case
In order to predict the system performance, the model is (without LHS unit), the daily specific cooling production and solar
implemented in a computer program developed in FORTRAN. This coefficient of performance are found to be DSCP0 ¼ 1626 kJ/kg and
allows computing the temperature, pressure and adsorbed mass, SCOP0 ¼ 0.136 with a number of refrigeration cycles of about
 A. El Fadar / Energy 114 (2016) 10e23 19

Table 3
System performance according to latent heat storage efficiency, W ¼ 1 m; Lr ¼ 0.5 m.

LHS Recovered Time at the end Number of Daily specific Coefficient of SCOP increase in
efficiency, h (%) heat, Qrec (kJ) of cooling production, refrigeration cooling production, performance, comparison with
t3 or t2 (hr) cycles, N () DSCP (kJ/kg) SCOP (%) 2nd case (%)

1st case: with LHS 50 3693 t3 ¼ 24.07 31.30 z 31 2062 17.30 þ26.83
70 5170 t3 ¼ 25.18 33.95 z 34 2237 18.76 þ37.53
90 6647 t3 ¼ 26.28 36.60 z 37 2411 20.22 þ48.24
2nd case: without LHS e e t2 ¼ 21.31 24.68 z 25 1626 13.64 e

N0 z 25 (see Table 3).
In case of using LHS unit, as pointed out above, when temper-
ature in HWST exceeds the suitable driving source temperature,
heat could be retrieved from HWST and stored in LHS unit. From the
current simulation, the saved thermal energy is found to be
Qsav ¼ 7385.79 kJ. As for the recovered heat, it depends on latent
heat storage efficiency (h). Thus, the different results obtained are
reported in Table 3, namely: the values of recovered heat (Qrec),
number of refrigeration cycles (N1), daily specific cooling produc-
tion (DSCP1) and solar coefficient of performance (SCOP1) in func-
tion of the efficiency h ranging from 0.5 to 0.9. These results are
discussed through three possible scenarios as follows:
increased by 48.24% in comparison with the process without
LHS.
- With a “reasonable” scenario (h ¼ 0.70), the DSCP1 and SCOP1
are found to be 2237 kJ/kg and 0.187 respectively. Qrec is esti-
mated to 5170 kJ, which corresponds approximately to N1 z 34
cooling cycles that could be accomplished. This means that the
system could provide cooling continuously up to t3 ¼ 25.18 h
(t3 z 01:11 a.m.). The SCOP is improved by 37.53% when
compared with the process without LHS.
5.2. Influence of the collector area on the system performance

- With a “pessimistic” scenario, assuming that only half of saved
heat is recovered from PCM unit (h ¼ 0.5), the DSCP1 and SCOP1
are found to be 2062 kJ/kg and 0.173 respectively. Qrec is esti-
mated to 3693 kJ, which could raise the number of cooling cycles
to nearly N z 31, that is to say, the system could furnish cooling
continuously until t3 ¼ 24.07 h with an increase in SCOP of
around 26.83% compared with the process without LHS.
- In an “optimistic” scenario (h ¼ 0.9), Qrec is estimated to 6647 kJ
and the cooling production could be extended up to t3 ¼ 26.28 h
(t3 z 02:17 a.m.) with an increase in number of cooling cycles
from N z 25 to roughly N z 37. The DSCP1 and SCOP1 are
evaluated to 2411 kJ/kg and 0.202 respectively. The SCOP is
Besides, the system performance depends strongly on solar
irradiance collected by the concentrator. To quantitatively elucidate
this issue, we have studied the effect of the aperture width of col-
lector (W) on the temperature of storage tank and on system per-
formance; the latent heat storage efficiency is kept constant:
h ¼ 0.7. The results are summarized in Table 4 and Fig. 5. Firstly,
Fig. 5a and Table 4a show that neither LHS unit nor SHS tank could
be useful unless the PTC area (Lc  W) is higher than (1.5  0.6 m2)
because, when the aperture width of collector W ¼ 0.6 m, the saved
thermal energy (Qsav ¼ 265 kJ), recovered heat (Qrec ¼ 185 kJ), time
of cooling production without (t2 e t1 ¼ 2.29 h) and with LHS unit
(t3 e t2 ¼ 0.15 h), DSCP (385 kJ/kg) and SCOP (z5%) are
insignificant.

Table 4
System performance in function of aperture width of the collector, h ¼ 0.7; Lr ¼ 0.5 m: (4a) W ¼ 0.6 m; (4b) W ¼ 1.2 m; (4c) W ¼ 1.4 m.

(a): W ¼ 0.6 m ; Qsav ¼ 265 kJ

LHS Recovered Time at the end of cooling Number of Daily specific cooling Coefficient of SCOP increase in
efficiency, heat, production, t3 or t2 (hr) refrigeration production, DSCP (kJ/kg) performance, SCOP (%) comparison with
h (%) Qrec (kJ) cycles, N () 2nd case (%)

1st case: 70 185 t3 ¼ 16.15 5.8 z 6 385 5.38 þ5.9

with LHS
2nd case: e e t2 ¼ 16 5.5 z 5 363 5.08 e
without LHS

(b): W ¼ 1.2 m ; Qsav ¼ 11956 kJ

LHS Recovered Time at the end of cooling Number of Daily specific cooling Coefficient of SCOP increase in
efficiency, heat, production, t3 or t2 (hr) refrigeration production, DSCP (kJ/kg) performance, SCOP (%) comparison with
h (%) Qrec (kJ) cycles, N () 2nd case (%)

1st case: 70 8369 t3 ¼ 30.02 47.09 z 47 3102 21.68 þ46.78

with LHS
2nd case: e e t2 ¼ 23.76 32.08 z 32 2113 14.77 e
without LHS

(c): W ¼ 1.4 m ; Qsav ¼ 16750 kJ

LHS Recovered Time at the end of cooling Number of Daily specific cooling Coefficient of SCOP increase in
efficiency, heat, production, t3 or t2 (hr) refrigeration production, DSCP (kJ/kg) performance, SCOP (%) comparison with
h (%) Qrec (kJ) cycles, N () 2nd case (%)

1st case: 70 11,725 t3 ¼ 34.86 59.8 z 60 3915 23.61 þ54.21

with LHS
2nd case: e e t2 ¼ 26.10 38.8 z 39 2556 15.31 e
without LHS
20
Fig. 5. Temperature evolution within storage tank with and without heat removal versus time, h ¼ 0.7 and Lr ¼ 0.5 m: (5a) W ¼ 0.6 m; (5b) W ¼ 1.2 m; (5c) W ¼ 1.4 m.
Fig. 5 (a, b, c) and Table 4 (a, b, c) show that the time of cooling
production and system performances (DSCP, SCOP) increase with
the aperture width of collector: the DSCP and SCOP increase from
385 to 3915 kJ/kg and from 5.38% to 23.61% respectively, when W
varies from 0.6 to 1.2 m. With W ¼ 1.4 m, the SCOP is improved by
54.21% in comparison with the process without LHS.
5.3. Effect of adsorbent bed mass on the system performance
To illustrate the influence of adsorbent bed mass on the system
performance, we have reported the simulation results in Figs. 6e8
which depict the effect of adsorbent bed mass (mAC), by varying the
adsorber length, on daily specific cooling production, number of
cooling cycles and solar coefficient of performance, respectively, in
the two cases: with and without LHS. Thus, it is shown from Fig. 6 (a
and b) that both DSCP0 and DSCP1 decrease considerably when mAC
increases from 3.3 to 13 kg when aperture width of collector is
W ¼ 0.8 m and to 21 kg when W ¼ 1.6 m. Beyond these values of
A. El Fadar / Energy 114 (2016) 10e23
mAC, the DSCP0 and DSCP1 remain invariable and insufficient (be-
tween 20 and 30 kJ/kg). This decrease of daily cooling production
could be interpreted as follows: since an adsorber with high
adsorbent bed mass absorbs more thermal energy and the share of
heat required to desorb a large amount of refrigerant becomes
insufficient which results in decrease of number of cooling cycles,
as shown from Fig. 7 (a and b).
Fig. 8 (a and b) shows the effect of mAC on solar coefficient of
performance. It can be observed that, for slight values of mAC
(mAC < 7 kg when W ¼ 0.8 m, and mAC < 21 kg when W ¼ 1.6 m),
the SCOP0 and SCOP1 fluctuate slightly around a certain value
(SCOP0 z 0.12 and SCOP1 z 0.14 for W ¼ 0.8 m, and SCOP0 ¼ 0.22
and SCOP1 ¼ 0.26 for W ¼ 1.6 m). Beyond theses ranges of mAC, the
SCOP0 and SCOP1 become constant and negligible (2%), because,
with high values of mAC, a considerable amount of thermal energy
is used to heat up the components of adsorber (sensible heat), as
result a small amount of refrigerant is desorbed, which means that
a small amount of cooling is produced and weak values of solar
 A. El Fadar / Energy 114 (2016) 10e23 21

Adsorber length (m) Adsorber length (m)

(a) 1 2 3 4 5 6 (b) 1 2 3 4 5 6
1400
6000
Daily specific cooling production (kJ/ kg)

Daily specific cooling production (kJ/ kg)

1200 DSCP0 (without LHS) DSCP0 (without LHS)
DSCP1 (with LHS) DSCP1 (with LHS)
5000
1000
4000
800

3000
600

2000
400

200 1000

0 0
5 10 15 20 25 30 35 40 5 10 15 20 25 30 35 40
Adsorbent bed mass (kg) Adsorbent bed mass (kg)

Fig. 6. Effect of adsorbent bed mass on daily specific cooling production, h ¼ 0.7: (6a) W ¼ 0.8 m; (6b) W ¼ 1.6 m.

Adsorber length (m) Adsorber length (m)

(a) 1 2 3 4 5 6
(b) 1 2 3 4 5 6
25

80
N0 (without LHS) N0 (without LHS)
20 N1 (with LHS) N1 (with LHS)
Number of cooling cycles (-)
Number of cooling cycles (-)

60
15

40
10

5 20

0 0

5 10 15 20 25 30 35 40 5 10 15 20 25 30 35 40
Adsorbent bed mass (kg) Adsorbent bed mass (kg)

Fig. 7. Effect of adsorbent bed mass on number of cooling cycles, h ¼ 0.7: (7a) W ¼ 0.8 m; (7b) W ¼ 1.6 m.

coefficient of performance are obtained. It is clear from Fig. 8 that additional thermophysical properties of PCM and HTF.
values of mAC greater than 7 kg and 21 kg for W ¼ 0.8 m and In the current work, from simulation results shown for example
W ¼ 1.6 m respectively are not recommended. It is obvious as well in Fig. 5b, when W ¼ 1.2 m, h ¼ 0.7 and Lr ¼ 0.5 m, the charging
from Figs. 6e8, that values of mAC higher than 5 kg and 10 kg for cycle may occur between t1 ¼ 10.40 h and t2 ¼ 23.76 h i.e. during
W ¼ 0.8 m and W ¼ 1.6 m respectively are not recommended in Dt1 ¼ 13.36 h, while the discharging process may happen between
case of integration of a LHS unit, since the performance of system t2 ¼ 23.76 h and t3 ¼ 30.02 h i.e. during Dt2 ¼ 6.26 h. These times
with LHS coincides with that of system without LHS. This means (Dt1 and Dt2) could be sufficient for storing and releasing the
that heat accumulated in HWST is insufficient to be retrieved and required heat. However, there are some techniques that should be
stored in LHS unit. taken into account to enhance the rate of heat transfer between the
In spite of the benefits of latent heat storage, mentioned above, PCM and the HTF and hence to shorten the response time. For
it presents the drawback of slow dynamic thermal response mainly example, inorganic PCMs (e.g. salt hydrates) could be chosen which
due to the low thermal conductivity of PCM, which is a major present the advantage of higher thermal conductivity (roughly
limitation in this field although the numerous investigations that 0.5 W/m K) in comparison with organic PCMs (around 0.2 W/m K)
were achieved to enhance the thermal conductivity. As result, an [52]. Furthermore, the encapsulation type plays a role in response
increase in charging and discharging time. Thus, the response time time: the PCM micro-encapsulated material releases its energy
for charging and discharging is an essential issue that should be more rapid during solidification process, contrary to the macro-
taken into account when designing a LHS unit, which is obviously encapsulated PCM which contributes to a slower solidification
complex, since the thermal response is affected by several condi- process [53]. The main thermophysical properties of some PCMs
tions, such as design and operating parameters of LHS unit and that could be used in proposed system are listed in Table 5.
22 A. El Fadar / Energy 114 (2016) 10e23

Adsorber length (m) Adsorber length (m)

(a) 1 2 3 4 5 6
(b) 1 2 3 4 5 6

0.3

SCOP0 (without LHS) 0.5 SCOP0 (without LHS)

Solar coefficient of performance (-)

Solar coefficient of performance (-)

SCOP1 (with LHS) SCOP1 (with LHS)

0.4
0.2

0.3

0.1
0.2

0.1
0
0

5 10 15 20 25 30 35 40 5 10 15 20 25 30 35 40
Adsorbent bed mass (kg) Adsorbent bed mass (kg)

Fig. 8. Influence of adsorbent bed mass on solar coefficient of performance, h ¼ 0.7: (8a) W ¼ 0.8 m; (8b) W ¼ 1.6 m.

Table 5
Thermophysical properties of salt hydrate materials (S) that could be used for latent heat storage unit [54,55].

PCM product Melting point (
C) Latent heat (kJ/kg) Thermal conductivity (W/m
C) Density (kg/m3) Specific heat (kJ/kg
C)

S89 89 151 0.67 1555 2.48

S83 83 141 0.62 1600 2.31

Besides, it is worthy to note that the results shown above have increased from roughly 25 to 37, and from 1626 to 2411 kJ/kg
been obtained under mild weather conditions of Tetouan, a Medi- respectively. The time of continuous cooling effect is
terranean city situated in the extreme north of Morocco. Thus, the extended up to t3 ¼ 26.28 h and the solar coefficient of
proposed process could be more interesting in sunnier and warmer performance is improved by 48.24%.
regions where the cooling needs are higher, e.g. regions in south of iii) The use of thermal energy storage, LHS or SHS, could be
Morocco as can be seen in Fig. 2. useful only if the collector area (Lc  W) is higher than
In light of this discussion, thanks to the thermal energy man- (1.5  0.6 m2). The results revealed also that the system
agement undertaken above, discernible improvements of the sys- performance increases with the aperture width of collector.
tem performance could be obtained; the intermittence problem iv) Under the simulation conditions, only values of adsorbent
could widely be overcome as well. Consequently, the proposed bed mass lower than 5 kg and 10 kg for W ¼ 0.8 m and
process seems to be an effective approach for cooling purposes in W ¼ 1.6 m respectively are recommended in order to make
both building and industrial sectors. the LHS unit useful.

6. Conclusions and outlook Finally, depending upon the magnitude and kind of the cooling
application, further improvements of the system performance
The goal of the current paper was to present a new cooling should be accomplished with an optimal configuration, taking into
process for solar continuous adsorption refrigeration system. It is account the economic considerations. Such a study could be carried
mainly based on combination of two (latent and sensible) heat out in a future work.
storage units and two adsorbent beds with a parabolic trough
collector. It has been proven numerically that the proposed process Appendix A. Supplementary data
could be a promising approach in thermal energy management for
saving energy, overcoming the intermittence problem, and Supplementary data related to this article can be found at http://
achieving performance enhancement. Based on the numerical dx.doi.org/10.1016/j.energy.2016.07.149.
investigation results, the main findings are summarized as follows:

i) When only one sensible heat storage tank is used, the daily References
specific cooling production and solar coefficient of perfor- [1] International Energy Agency. Technology roadmap. Solar heating and cooling.
mance are found to be 1626 kJ/kg and 0.13 respectively; 2012. http://www.iea-shc.org/data/sites/1/publications/2012_
cooling could be produced continuously but only in the time SolarHeatingCooling_Roadmap.pdf.
[2] International Energy Agency. Energy Technology Perspectives. Mobilising
range from t1 ¼ 11.03 to t2 ¼ 21.31 h, with a number of
innovation to accelerate climate action, 2015. 2015.
refrigeration cycles per day of about 25. [3] El Fadar Abdellah. Thermal behavior and performance assessment of a solar
ii) With the same aperture width of collector (W ¼ 1 m), when a adsorption cooling system with finned adsorber. Energy 2015;83:674e84.
second latent heat storage unit with efficiency of h ¼ 90% is [4] Balaras CA, Grossman G, Henning HM, Ferreira CAI, Podesser E, Wang L, et al.
Solar air conditioning in Europedan overview. Renew Sust Energy Rev
integrated into the system, the number of refrigeration cycles 2007;11(2):299e314.
per day and daily specific cooling production are significantly [5] Choudhury B, Saha BB, Chatterjee PK, Sarkar JP. An overview of developments
 A. El Fadar / Energy 114 (2016) 10e23
in adsorption refrigeration systems towards a sustainable way of cooling. Appl
Energy 2013;104:554e67.
[6] Mahesh A, Kaushik SC. Solar adsorption cooling system: an overview. J Renew
Sustain Energy 2012;4(022701).
[7] Wang RZ, Oliveira RG. Adsorption refrigerationean efficient way to make
good use of waste heat and solar energy. Prog Energy Combust Sci 2006;32:
424e58.
[8] Solmuş I_ Rees DAS, Yamalı C, Baker D, Kaftanoglu B. Numerical investigation of
coupled heat and mass transfer inside the adsorbent bed of an adsorption
cooling unit. Int J Refrig 2012;35:652e62.
[9] Li XH, Hou XH, Zhang X, Yuan ZX. A review on development of adsorption
coolingdNovel beds and advanced cycles. Energy Convers Manag 2015;94:
221e32.
[10] El Fadar A, Mimet A, Pe rez-García M. Study of an adsorption refrigeration
system powered by parabolic trough collector and coupled with a heat pipe.
Renew Energy 2009;34:2271e9.
[11] Tian Y, Zhao CY. A review of solar collectors and thermal energy storage in
solar thermal Applications. Appl Energy 2013;104:538e53.
[12] Zhang XJ, Wang RZ. Design and performance simulation of a new solar
continuous solid adsorption refrigeration and heating hybrid system. Renew
Energy 2002;27:401e15.
[13] Alghoul MA, Sulaiman MY, Sopian K, Azmi BZ. Performance of a dual-purpose
solar continuous adsorption system. Renew Energy 2009;34(3):920e7.
[14] El Fadar A, Mimet A, Pe rez-García M. Modelling and performance study of a
continuous adsorption refrigeration system driven by parabolic trough solar
collector. Sol Energy 2009;83:850e61.
[15] Zhai XQ, Wang RZ. Experimental investigation and performance analysis on a
solar adsorption cooling system with/without heat storage. Appl Energy
2010;87:824e35.
[16] Li C, Wang RZ, Wang LW, Li TX, Chen Y. Experimental study on an adsorption
icemaker driven by parabolic trough solar collector. Renew Energy 2013;57:
223e33.
[17] Abu-Hamdeh Nidal H, Alnefaie Khaled A, Almitani Khalid H. Design and
performance characteristics of solar adsorption refrigeration system using
parabolic trough collector: experimental and statistical optimization tech-
nique. Energy Convers Manag 2013;74:162e70.
[18] El-Sharkawy Ibrahim I, AbdelMeguid Hossam, Saha Bidyut Baran. Potential
application of solar powered adsorption cooling systems in the Middle East.
Appl Energy 2014;126:235e45.
[19] Deshmukh H, Maiya MP, Murthy SS. Continuous vapour adsorption cooling
system with three adsorber beds. Appl Therm Eng 2015;82:380e9.
[20] Xu Biwan, Li Zongjin. Paraffin/diatomite/multi-wall carbon nanotubes com-
posite phase change material tailor-made for thermal energy storage cement-
based composites. Energy 2014;72:371e80.
[21] Aadmi Moussa, Karkri Mustapha, El Hammouti Mimoun. Heat transfer char-
acteristics of thermal energy storage for PCM (phase change material) melting
in horizontal tube: numerical and experimental investigations. Energy
2015;85:339e52.
[22] Zhang HL, Baeyens J, Degre ve J, Ca
ceres G, Segal R, Pitie
 F. Latent heat storage
with tubular-encapsulated phase change materials (PCMs). Energy 2014;76:
66e72.
[23] Laing D, Bauer T, Breidenbach N, Hachmann B, Johnson M. Development of
high temperature phase-change-material storages. Appl Energy 2013;109:
497e504.
[24] Naghavi MS, Ong KS, Mehrali M, Badruddin IA, Metselaar HSC. A state-of-the-
art review on hybrid heat pipe latent heat storage systems. Energy Convers
Manag 2015;105:1178e204.
[25] Khadiran Tumirah, Hussein Mohd Zobir, Zainal Zulkarnain, Rusli Rafeadah.
Encapsulation techniques for organic phase change materials as thermal en-
ergy storage medium: a review. Sol Energy Mater Sol Cells 2015;143:78e98.
[26] Soares N, Costa JJ, Gaspar AR, Santos P. Review of passive PCM latent heat
thermal energy storage systems towards buildings' energy efficiency. Energy
Build 2013;59:82e103.
[27] Tamainot-Telto Z, Metcalf SJ, Critoph RE, Zhong Y, Thorpe R. Carbon-
eammonia pairs for adsorption refrigeration applications: ice making, air
conditioning and heat pumping. Int J Refrig 2009;32(6):1212e29.
[28] Srivastava NC, Eames IW. A review of adsorbents and adsorbates in solide
vapour adsorption heat pump systems. Appl Therm Eng 1998;18:707e14.
[29] Cacciola G, Restuccia G. Reversible adsorption heat pump: a thermodynamic
23
model. Int J Refrig 1995;18(2):100e6.
[30] Liu YL, Wang RZ, Xia ZZ. Experimental study on a continuous adsorption
water chiller with novel design. Int J Refrig 2005;28:218e30.
[31] Abedin AH, Rosen MA. Closed and Open thermochemical energy storage:
energy- and exergy-based comparisons. Energy 2012;41:83e92.
[32] Hongois S, Kuznik F, Stevens P, Roux JJ. Development and characterization of a
new MgSO4-zeolite composite for long-term thermal energy storage. Sol
Energy Mater Sol Cells 2011;95:1831e7.
[33] Ristic A, Maucec D, Henninger SK, Kaucic V. New two-component water
sorbent CaCl2-FeKIL2 for solar thermal energy storage. Microporous Meso-
porous Mater 2012;164:266e72.
[34] National Agency for the Development of Renewable Energy and Energy Effi-
ciency (ADEREE), renewable energy atlas. www.aderee.ma/atlas-ER.
[35] Kousksou T, Allouhi A, Belattar M, Jamil A, El Rhafiki T, Zeraouli Y. Morocco's
strategy for energy security and low-carbon growth. Energy 2015;84:98e105.
[36] Dubinin MM, Astakhov VA. Development of the concept of volume filling of
micropores in the adsorption of gases and vapors by microporous adsorbents.
Washington, DC, USA: American Chemical Society; 1971.
[37] Lane GA. Solar heat storage: latent heat material, volume I: background and
scientific principles. Florida: CRC Press; 1983.
[38] Zhao CY, Ji Yunan, Xu Zhiguo. Investigation of the Ca(NO3)2eNaNO3 mixture
for latent heat storage. Sol Energy Mater Sol Cells 2015;140:281e8.
[39] Sharma Atual, Tyagi VV, Chen CR, Buddhi D. Review on thermal energy
storage with phase change materials and applications. Renew Sustain Energy
Rev 2009;13:318e45.
[40] Kenisarin Murat. Khamid Mahkamov. Salt hydrates as latent heat storage
materials:Thermophysical properties and costs. Sol Energy Mater Sol Cells
2016;145:255e86.
[41] Chiu Justin NW, Martin Viktoria. Multistage latent heat cold thermal energy
storage design analysis. Appl Energy 2013;112:1438e45.
[42] Wang Weilong, Li Hailong, Guo Shaopeng, He Shiquan, Ding Jing, et al. Nu-
merical simulation study on discharging process of the direct-contact phase
change energy storage system. Appl Energy 2015;150:61e8.
[43] Kuravi S, Trahan J, Goswami DY, Rahman MM, Stefanakos EK. Thermal energy
storage technologies and systems for concentrating solar power plants. Prog
Energy Combust Sci 2013;39:285e319.
[44] The International Renewable Energy Agency (IRENA). Thermal energy storage.
2013. http://www.irena.org/publications.
[45] Hauer A. Storage technology issues and opportunities, committee on energy
research and technology (International Energy Agency), international low-
carbon energy technology platform, strategic and cross-cutting workshop
“energy storagee issues and opportunities”. 15 February 2011. Paris. France.
[46] Chai Luxiao, Wang Xiaodong, Wu Dezhen. Development of bifunctional
microencapsulated phase change materials with crystalline titanium dioxide
shell for latent-heat storage and photocatalytic effectiveness. Appl Energy
2015;138:661e74.
[47] El Mghouchi Y, El Bouardi A, Choulli Z, Ajzoul T. New model to estimate and
evaluate the solar radiation. Int J Sustain Built Environ 2014;3:225e34.
[48] Hassan HZ, Mohamad AA, Bennacer R. Simulation of an adsorption solar
cooling system. Energy 2011;36:530e7.
[49] Allouhi A, Kousksou T, Jamil A, Zeraouli Y. Modeling of a thermal adsorber
powered by solar energy for refrigeration applications. Energy 2014;75:
589e96.
[50] El Fadar A, Mimet A, Azzabakh A, Pe rez-García M, Castaing J. Study of a new
solar adsorption refrigerator powered by a parabolic trough collector. Appl
Therm Eng 2009;29:1267e70.
[51] Osborne Nathan S, Van Dusen Milton S. Specific heat of liquid ammonia.
National Bureau of Standards. Bull Bureau Stand 1918;14:397e432.
[52] Arteconi A, Hewitt NJ, Polonara F. State of the art of thermal storage for
demand-side management. Appl Energy 2012;93:371e89.
[53] Porteiro J, Míguez JL, Crespo B, Gonza lez LML, De Lara J. Experimental
investigation of the thermal response of a thermal storage tank partially filled
with different PCMs (phase change materials) to a steep demand. Energy
2015;91:202e14.
[54] PCM products, Available at: (http://www.pcmproducts.net/Phase_Change_
Material_Products.html).
[55] Kenisarin Murat. Khamid Mahkamov. Salt hydrates as latent heat storage
materials:Thermophysical properties and costs. Sol Energy Mater Sol Cells
2016;145:255e86.