Applied Thermal Engineering 87 (2015) 185e199

Contents lists available at ScienceDirect

Applied Thermal Engineering

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

Research paper

Simulation study of a combined adsorption refrigeration system

Alireza Sadeghlu a, Mortaza Yari b, *, Hossein Beidaghy Dizaji a
a
Young Researchers and Elite Club, Tabriz Branch, Islamic Azad University, Tabriz, Iran
b
Faculty of Mechanical Engineering, University of Tabriz, Tabriz, Iran

h i g h l i g h t s

Development of a thermodynamic model for a combined adsorption refrigeration system.

Zeolite 13X/CaCl2 is considered as a new adsorbent in a combined adsorption refrigeration system.

Zeolite 13X/CaCl2-water are more advantageous than the other working pairs.

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

Article history: In this study, we have divided combined ADRS into four types based on different arrangements of two
Received 8 December 2014 working pairs, Zeolite 13x/CaCl2-water and Silica gel (RD type)ewater, to analyze the performance of
Accepted 7 May 2015 combined ADRS. After validating mathematical models with available experimental data, ADRS is
Available online 19 May 2015
simulated by using SimulinkeMatlab software to achieve optimum times for various processes. The
results of simulation show that the cooling capacity of the system with Zeolite 13x/CaCl2-water is more
Keywords:
than the other types. The results have shown that the arrangement of adsorbents affects cooling capacity
Adsorption refrigeration
of combined ADRS significantly. In Type A, Zeolite 13x/CaCl2-water has been used as an adsorbent for
Cascade cycle
Simulation
both top and bottom cycles. This type not only has more cooling capacity than the other types, but also
the effect of hot water temperature on cooling capacity of this type is less than the others. Furthermore, a
sensitivity analysis has been done to determine the importance of each parameter on ADRS system
because the cooling capacity and the COP are influenced by many constant parameters.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction simulated models. Boalman et al. [5] experimentally have analyzed

The necessity of using renewable energy sources and environ-
mental friendly refrigerants is increasingly recognized due to the
ozone layer depletion and global warming. In the recent years,
various technologies have been developed to make best use of
renewable energies; ADRS is one of these technologies [1]. This
system, in comparison with conventional vapor compression sys-
tems, which consumes less electrical energy and is capable of
storing energy when it is run by solar or industrial waste energy.
Better control and less vibration can be accounted as advantages of
this system. Furthermore, the operating and maintenance cost of
ADRS is less than that of the vapor compression cooling system [2].
The researches which have been done on adsorption refrigeration
systems can be categorized in three main groups: First we see
researchers such as Saha et al. [3], Chua et al. [4] who presented
* Corresponding author. Tel.: þ98 41 33356026.
E-mail address: myari@Tabrizu.ac.ir (M. Yari).
the effects of various parameters on performance of simple cycle of
ADRS (two-bed ADRS). Mahdavikhah and Niazmand numerically
have studied the effects of geometric parameters of heat exchanger
on the adsorption chiller performance. They have used a three-
dimensional model to predict the dynamic performance of an
intermittent adsorption cycle working with the composite sorbent
SWS-1L and water. The results show that COP increases as the bed
height increases. It is less sensitive to the fins spacing, though.
Furthermore, the SCP decreases dramatically with increasing fin
height due to the increase in cycle time and adsorbent mass.
Therefore, the geometric specifications of the adsorption heat
exchanger are of great practical importance in the design of the
adsorption cooling systems [6]. Niazmand et al. has shown that
optimum diameter is influenced by different parameters such as
fins height and spacing and heating/cooling temperatures of the
thermal fluid. They have show that the COP of system remains
almost constant for the selected range of particle sizes at fixed
geometrical and operational conditions; whereas, SCP of system
has its maximum value at a specific particle diameter [7]. Another

http://dx.doi.org/10.1016/j.applthermaleng.2015.05.009
1359-4311/© 2015 Elsevier Ltd. All rights reserved.
186 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199

Table 1
Different type of adsorbents which are used in simulation of combined ADRS.

Down cycle adsorbents Upper cycle adsorbents

Type SE1 SE2 SE3 SE4

A Zeolite 13X/CaCl2 Zeolite 13X/CaCl2 Zeolite 13X/CaCl2 Zeolite 13X/CaCl2

B Zeolite 13X/CaCl2 Zeolite 13X/CaCl2 Silica gel Silica gel
(RD type) (RD type)
C Silica gel Silica gel Zeolite 13X/CaCl2 Zeolite 13X/CaCl2
(RD type) (RD type)
D Silica gel Silica gel Silica gel Silica gel
(RD type) (RD type) (RD type) (RD type)

Table 2
Numerical constants of equations (1)e(4) [3,15].

Parameter subscript 0 1 2 3

A 6.5314 0.072452 0.23951  103 0.25493  106

B 15.587 0.15915 0.50612  106 0.5329  106
C- Adsorption 381.4 3.463 10.08  103 9.153  106
C- Desorption 400.8 5.532 10.02  103 8.908  106

Table 3
Values of parameters adopted in the simulation [15,26].

Symbol Value Unit

(UA)cond 4115.23  3.73 W/K

(UA)evap 2557.54  1.91 W/K
(UA)bed 1724.14  2.46 W/K
Mcond 24.28 kg
Mevap 12.45 kg
Ma 47 kg
Mcond,ref 20 kg
Mevap,ref 50 kg
Qis,RD silica gel 2800 kJ/kg
Qis,zeolite 3394.48 kJ/kg
hfg 2500 kJ/kg
Rp,RD silica gel 1.7  104 m
Rp,zeolite 1  106 m
Mf,Hex 64.04 kg
Mk,Hex 51.20 kg
Cp,w 4.186 kJ/kg K
Cp,Al 0.905 kJ/kg K
Fig. 1. Schematic configuration of combined ADRS. Cp,Cu 0.386 kJ/kg K
Cp,RD silica gel 0.924 kJ/kg K
Cp,zeolite 0.836 kJ/kg K
Ea,RD silica gel 4.2  104 kJ/kmol
Ea,zeolite 2.8035  104 kJ/kmol
Dso,RD silica gel 2.54  104 m2/s
Dso,zeolite 3.92  106 m2/s

Table 4
Nominal operating condition for the combined ADRS.

Parameter Symbol Value Unit


Hot water inlet temperature Thw,in 75 C
Hot water mass flow rate m_ hw 0.65 kg/s

Adsorber and condenser cooling water Tcw,(ads,cond),in 30 C
inlet temperature
Adsorber and condenser cooling water m_ cw;adsþcond (0.65 þ 0.65) kg/s
mass flow rate

Chilled water inlet temperature Tchill,in 15 C
Chilled water mass flow rate m_ chill 0.6 kg/s

group of researchers analyzed the performance of different adsor-

Fig. 2. Schematic configuration of combined evaporator-condenser unit.
bents and refrigerants on ADRS such as Silica gel (A type)-water [8],
Silica gel (RD type)-water [3,5], Zeolite-water [9] and Zeolite 13x-
water [10,11]. Furthermore, these researchers enriched solid
adsorbent with mineral salts such as CaCl2, LiBr, and etc. So, they
were able to increase both adsorption capacity and performance of
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199
Fig. 3. Simulation model of the combined ADRS.
Fig. 4. Simulated and experimental outlet temperature profiles of heat transfer fluids
for the conventional Silica gel (RD type)-water ADRS.
simple ADRS [12e17]. The third group analyzed various structures
of ADRS to improve its performance. Among them, one can refer to
Saha et al. [18,19]. In 1997 and 2006 they assessed the adsorption
refrigeration system with low temperature energy sources (be-
tween 40  C and 95  C) in two different modes. First they studied
multi-bed regenerative mode in which the temperature of heat
energy sources was 60  C to 95  C. Second, they studied multi-stage
regenerative mode with heat energy temperature sources between
40  C and 60  C. Although, ADRS dual mode changed the perfor-
mance of simple ADRS acceptably, the results of researches have
showed that irreversible heat losses increase due to the fact that a
number of switching process between adsorbers and desorbers
increases in dual mode ADRS. Song et al. have studied experi-
mentally a system in which they combined adsorption freezing
187
Fig. 5. The effect of switching time on cooling capacity for Silica gel (RD type)-water
ADRS.
cycle and adsorption air conditioning cycle. In that study they have
connected two systems in a way that they could use the outlet heat
source of adsorption freezing cycle as an inlet heat source of
adsorption air conditioning cycle [20]. Habib et al. [21] have
analyzed the performance of combined ADRS (with 70  C energy
source) in which a heat exchanger is used not only as an evaporator,
but also as a condenser at the same time. By Using R134a and
activated carbon adsorbent for high cycle and activated carbon
adsorbent R507a refrigerant for low cycle, they reached the result
of 85 kW cooling capacity with 10  C chilled water outlet tem-
perature, and 0.069 coefficient of performance. Habib et al. [22]
have analyzed the same combined adsorption system as their
previous work [21], but they used hot water outlet temperature of
one cycle as a heat source of desorption and pre-heating processes
188 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199
Fig. 6. The effect of switching time on COP for both experimental data [27] and
simulation of Zeolite 13x/CaCl2-water ADRS.
of the other cycle. So, they reported 65.2 kW cooling capacity with
0.106 coefficient of performance, and chilled water outlet
temperature 10  C with 76  C heat source.
One of the reasons which limit two-bed ADRS system is its low
performance. Although the performance of two-bed ADRS has just
been improved by composite adsorbent [16], the performance of
the system needs more researches to be done. Therefore, it seems
necessary to use different structures of ADRS and new adsorbents
to increase the performance of ADRS system.
There is no work has been done to study combined adsorption
refrigeration system of two working pairs: Zeolite 13x/CaCl2-water
and Silica gel (RD type)-water, and also combined adsorption
refrigeration system. In the present study, we attempt to assess the
performance of combined ADRS with two different working pairs.
One of them is Silica gel (RD type)-water and, the other one is
Zeolite 13x/CaCl2-water. Moreover, different experiments have re-
ported different experimental data for a specific material because
each of these experiments has been done in different conditions.
So, in this article, we have presented a sensitivity analysis to find
important parameters in order to determine their quantity with
double precision.
Fig. 7. Half cycle time effect on cooling capacity for different types of combined ADRS.
Fig. 8. Half cycle time effect on COP for different types of combined ADRS.
2. Combined ADRS description
A combined ADRS in Fig. 1 shows the consecutive composition
of two two-bed ADRSs. Shell and tube heat exchanger connect two
two-bed ADRS and act as an evaporator in higher cycle and a
condenser in lower cycle. The cycle is actually accomplished in two
stages: In the first stage, the sorption elements 1 and 3 (SE1 and
SE3) are considered as adsorbers, and the sorption elements 2 and 4
(SE2 and SE4) are considered as desorbers. In this stage, the valves
1, 3, 5 and 7 are opened and valves 2, 4, 6 and 8 are closed. At first,
absorbers (SE1 and SE3) are cooled by cooling water from cooling
tower; so, the vapors from the evaporators are adsorbed by the
adsorbents in the SE1 and SE3. At the same time, hot water from
thermal storage tank not only does heat the desorbers (SE2 and
SE4), but also evaporates refrigerants. The condensates then go into
evaporators via a pressure reducing U-bent tube and are absorbed
by SE1 and SE3 until the absorbents in the adsorbers are saturated.
In the second stage, SE2 and SE4 become adsorbers, and SE1 and
SE3 become desorbers. In this stage, the valves 2, 4, 6, and 8 are
opened, and 1, 3, 5, and 7 are closed. However, before this stage, the
pressures in SE1 and SE3 and SE2 and SE4 should be adjusted. So
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199 189

refrigeration circuits [23]. The evaporatorecondenser unite which

has been used in the present article is a shell-and-tube heat exchanger
as shown in Fig. 2. Engineering challenges of selecting the shell-and-
tube evaporator-condenser unit are mentioned as follow [24,6]:

Heat transfer efficiency (number of fins which are used to in-

crease the heat transfer between evaporator and condenser)

Choosing the best material to prevent corrosion problem

Saving time and money

Aligning the fins

The effects of fins spacing and height on the system performance

Volume is a big issue (evaporator-condenser unit should be
designed in compact from)

Noise should be reduced

Fig. 9. Switching time effect on cooling capacity for different types of combined ADRS. In the present study we attempt to analyze the performance of
the combined ADRS with different selected adsorbents according to
Table 1.
that SE1 and SE3 are heated (pre-heating) and SE2 and SE4 are
cooled (pre-cooling) when all the valves are closed. The sequence of
processes in the second stage is similar to the first one, but the 3. Mathematical modeling
arrangement of the processes operates inversely.
As well known, in a combined refrigeration system the two 3.1. Adsorption isotherms and kinetics
refrigeration circuits are thermally coupled through an intermediate
cascade heat exchanger, i.e. evaporator-condenser. The temperature In this paper, modified Freundlich (SeBK) equation is used to
difference in the cascade heat exchanger plays important roles in equilibrium water uptake of Silica gelewater pair [3].
determining the configuration and optimal performance of the

cascade system [23]. Currently, several types of cascade heat ex- Ps ðTw Þ BðTs Þ
xeq ¼ AðTs Þ (1)
changers such as plate, shell-and-plate or shell-and-tube heat ex- Ps ðTs Þ
changers can be employed for cascade systems to couple the two

Fig. 10. Temperature profiles of upper cycles for different components of Type A, Type B, Type C, Type D combined ADRS.
190 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199

Fig. 11. Temperature profiles of bottom cycles for different components of Type A, Type B, Type C, Type D combined ADRS.

where,
15Ds
Ks av ¼ (6)
R2P
AðTs Þ ¼ A0 þ A1 Ts þ A2 Ts2 þ A3 Ts3 (2)
In this equation, Rp is the adsorbent particle radius, and Ds is the
surface diffusivity which is defined by Arrhenius equation as
BðTs Þ ¼ B0 þ B1 Ts þ B2 Ts2 þ B3 Ts3 (3) below:

xeq With Kg/Kg unit shows for each Kg adsorbent how much Ea
refrigerant is adsorbed by adsorbent in equilibrium. Ps (Tw) is Ds ¼ Dso exp (7)
RT
saturation vapor pressure in refrigerant (water) temperature. Ps (Ts)
shows saturation vapor pressure in adsorbent (Silica gel) temper- where, Ea is activation energy, and R is the universal ideal gas
ature. The numerical values related to constant parameters are constant.
shown in Table 2. For the Zeolite 13X/CaCl2, the equilibrium water Using the principles of mass and energy conservations and the
uptake is correlated using the following equation [15]: relations obtained for the equilibrium water uptake as well as the
adsorption or desorption rate, the combined ADRS performance is
.  modeled.
xeq ¼ 1 C0 þ C1 Ta þ C2 Ta2 þ C3 Ta3 (4)

where, C0, C1, C2, and C3 are numerical constants; they are outlined 3.2. Energy balance
in Table 2.
Adsorption rate or kinetic of Zeolite 13X/CaCl2-water and Silica 3.2.1. adsorber and desorber
gel (RD type)-water pair is estimated by LDF model as given in Eq. In order to apply the lumped model for the performance of two-
(5) as below [25]: bed ADRS, the following assumptions has been made [26]:

dx   a) The temperature in each of the components i.e. the adsorber,

¼ Ks av xeq  x (5) desorber, condenser and evaporator remains the same through
dt
the whole component.
where, Ksav is the overall mass transfer coefficient for adsorbent/ b) The refrigerant vapor and heat losses to the environment are
refrigerant pair, and it can be written as: neglected.
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199 191

Fig. 12. Outlet temperature profiles of heat transfer fluids in upper cycles for Type A, Type B, Type C, Type D combined ADRS.

By considering these assumptions, the energy balance for 3.2.2. Condenser in higher cycle
adsorber and desorber can be written as below [3]. The equation of energy balance for condenser in higher cycle is
stated as:

 dTads=des
Ms xcpw þ Ms cps þ mAl CAl þ mCu CCu
dt d n 
up
o
Mcond;ref cp;w þ Mcond cp;Cu Tcond
dxads=des h   i dt
¼ 4Ms cpv Tads=des  Tevap=cond þ Qis
dx h  up  i
up
dt up
_ pw ðTin  Tout Þw;ads=des ¼ 4Ms des cpv Tdes  Tcond þ hfg
þ mc (8) dt
up
  dx
In this equation, 4 takes zero value in pre-cooling and pre-  mc_ p cw ðTout  T in Þcw þ 4cp;w T up Ma des (10)
cond dt
heating processes and unit value in adsorption and desorption
processes. The values for other parameters and the operating
conditions are observed in Tables 3 and 4. The subscripts i and j
The term at the left side of the equation represents the sensible
indicate the adsorption/desorption bed and cooling/heating sour-
heat required for the heat exchanger pipes, and condensed adsor-
ces, respectively. The first term on the left side of the Eq. (8) shows
bate. The first term on the right side of Eq. (10) indicates the effect
the change in internal energy of adsorbate and adsorbent. The
of adsorbate gaseous phase produced in desorption and generated
second term indicates the change in internal energy of heat ex-
heat during the condensation of adsorbat in condenser. Second
changers (including the pipes and fins). The first term on the right
term expresses the heat given to the cooling water and the last term
hand side of the Eq. (8) is related to the generated heat during
is the energy which is delivered to the middle evaporator via the U
adsorption process or the extracted heat in desorption process.
bend tube. The outlet water temperature of the condenser is
Finally, the second term on the right side of the Eq. (8) indicates the
measured by:
input heat to the cooling water during adsorption process or the
provided heat by hot water through the desorption process. " #
In order to assess the outlet temperature of heat transfer fluid in   ðUAÞcond
Tcw;out ¼ T cond þ Tcw;in þ T cond exp   (11)
adsorber or desorber, the log mean temperature difference model is _ p cond
mc
used:
" #
  ðUAÞbed
Tj;out ¼ Tjbed þ Tj;in  Tjbed exp  i (9)
_ p j
mc 3.2.3. Middle evaporator
The conservation of energy for middle evaporator is written as
follows:
192 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199

Fig. 13. Outlet temperature profiles of heat transfer fluids in bottom cycles for Type A, Type B, Type C, Type D combined ADRS.

d n 
m
o d n 
m
o
Mevap;ref cp;w þ Mevap cp;Cu Tevap Mcond;ref cp;w þ Mcond cp;Cu Tcond
dt dt
dx h  up  i dxdown h  down  i
up up
m up dx
¼ 4Ms ads cpv Tads  Tevap þ hfg  4cp;w Tcond Ma des ¼ 4Ms des m
cpv Tdes  Tcond þ hfg
dt dt dt
dxdown h   i dxdown dxup h  up  i
down m
 4Ms des cpv Tdes  Tcond þ hfg m
þ 4cp;w Tcond Ma des  4Ms ads cpv Tads m
 Tevap þ hfg
dt dt dt
up
m dxdown up dx
þ 4cp;w Tcond Ma des  4cp;w Tcond Ma des
dt dt
(12) (13)
The left term of Eq. (12) expresses the change of internal The left term of Eq. (13) expresses the change of internal
energies associated with the tubes, and adsorbate inside the energy associated with the tubes, and adsorbate inside the
middle evaporator. The first term on the right side of Eq. (12) middle evaporator. The first term on the right side of Eq. (12)
shows the effects of adsorbate gaseous phase produced in shows the effects of adsorbate gaseous phase due to desorption
adsorption, and latent heat of adsorbate. The second term is the process, and generated heat in condensation of adsorbate in
value of input energy from upper condenser to middle evapo- condenser. The second term is the value of energy delivered to
rator. The third term on the right indicates the effects of adsor- lower evaporator from middle condenser. The third shows the
bate gaseous phase produced in desorption, and generated heat effects of adsorbate gaseous phase due to adsorption process, and
during the condensation of adsorbate in condenser. Finally, the latent evaporation heat from adsorbate. Finally, the last term is
last term shows the energy delivered to lower evaporator from the value of input energy from upper condenser to middle
middle condenser. evaporator.

3.2.4. Middle condenser 3.2.5. Evaporator in lower cycle

The conservation of energy for middle condenser is written as The conservation of energy for evaporator in lower cycle is
follows: defined in the following equation:
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199 193

Fig. 14. Effect of cooling water inlet temperature on the cooling capacity at THot, in ¼ 55 Fig. 17. Effect of cooling water inlet temperature on the COP at THot, in ¼ 55 [ C].
[ C].

Fig. 15. Effect of cooling water inlet temperature on the cooling capacity at THot, in ¼ 75
[ C]. Fig. 18. Effect of cooling water inlet temperature on the COP at THot, in ¼ 75 [ C].

Fig. 16. Effect of cooling water inlet temperature on the cooling capacity at THot, in ¼ 85
[ C]. Fig. 19. Effect of cooling water inlet temperature on the COP at THot, in ¼ 85 [ C].
194 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199

Table 5

d n  o
Performance of different types of combined ADRS.
down
Type Switching time Adsorption/desorption time Cooling capacity COP
Mevap;ref cp;w þ Mevap cp;Cu Tevap
dt
A 30 500 21.38 0.2743 dxdown h  down down
 i
B 40 500 19.43 0.2663 ¼ 4Ms ads cpv Tads  Tevap þ hfg
C 40 400 13.51 0.2573
dt
D 40 400 12.25 0.2561   dxdown
 mc_ p chill ðTout  T in Þchill  4cp;w Tcond
m
Ma des (14)
dt
The first term at the left side of Eq. (14) expresses the change of
internal energy associated with the tubes and adsorbate inside the
evaporator. The effects of adsorbate gaseous phase due to

Fig. 20. Sensitivity analysis of cooling capacity and COP for Type A combined ADRS.

Fig. 21. Sensitivity analysis of cooling capacity and COP for Type B combined ADRS.
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199 195

adsorption process and latent evaporation heat of adsorbate is

cycle
shown by the first term on the right side. The second term is the Qchill
refrigeration capacity, and the last term is the value of energy
COP ¼ Upper
(19)
QHOT þ QHOT
Downer
flowing from middle condenser.

3.3. Conservation of mass 4. System simulation

The conservation of mass in the evaporator is: The combined ADRS performance is simulated by the Simulink-
Matlab software as shown in Fig. 3 in order to solve conservation
dmevap;ref dx dx equations of mass and energy in the adsorber, desorber, evaporator
¼ Ma des  Ma ads (15) and condenser.
dt dt dt

where,dxdes/dt and dxads/dt are desorption, and adsorption rates,
respectively.
3.4. The combined ADRS performance
For the combined ADRS performance, the cooling capacity and
coefficient of performance are expressed, respectively as:
5. Model validation
Before analyzing the effects of different parameters on com-
bined ADRS with different arrangements of working pairs: Silica gel
(RD type)-water and Zeolite 13x/CaCl2-water, the validity of
mathematical model and the accuracy of simulating equation
should be assessed. We have used two-bed ADRS data because
there is a lack of experimental data about combined system. For this
purpose, our simulation results are compared with available
Z tcycle  
experimental date of tow-bed ADRS in Figs. 4e6. It shows a good
_ p
mc chill
ðTin  T out Þchill dt
cycle 0 agreement between our present study and previous works [3,17,27]
Qchill ¼ (16)
tcycle in two-bed ADRS with two working pairs: Silica gel (RD type)-
water and Zeolite 13x/CaCl2-water. Therefore, we can use the
same simulation model to simulate combined ADRS, although there
Z
tcycle
Upper  
is no experimental data to compare with combined ADRS
QHOT ¼ _ p
mc hw
ðTin  T out Þhw dt (17) simulation.
0
6. Results and discussion
Z
tcycle
 
Fig. 7 shows the effect of half cycle (adsorption/desorption) time
Downer
QHOT ¼ _ p
mc hw
ðTin  T out Þhw dt (18)
on cooling capacity of different combined ADRSs based on Table 1
0
which working conditions have been specified in Table 3. As
shown in Fig. 7, the cooling capacity of Type A, Type B, Type C, and

Fig. 22. Sensitivity analysis of cooling capacity and COP for Type C combined ADRS.
196 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199
Type D are maximized at the half cycle time around 400s, 500s,
300s, and 500s respectively. In addition, when the half cycle time is
increasing, the adsorbents adsorb maximum amount of water va-
pors at optimum half cycle time. As a matter of fact, adsorbents do
not adsorb much water vapors after optimum half cycle time.
According to our previous study [16], the result shows that the
cooling capacity of Zeolite 13x/CaCl2-water two-bed ADRS is more
than the cooling capacity of Silica gel (RD type)-water two-bed
ADRS. Therefore, the cooling capacity of Type A combined ADRS is
more than the cooling capacity of Type D combined ADRS as shown
in Fig. 7. Similarly the cooling capacity of Type B is more than the
cooling capacity of Type C, as shown in Fig. 7, because the adsorbent
of bottom cycle has dominated effect on cooling capacity of com-
bined ADRS.
As shown in Fig. 8, for each combined ADRS Types, COP in-
creases as the half cycle time increases due to lower consumption of
heat energy in comparison with the cooling energy production
during the cycle time.
The effects of switching times on the cooling capacities are shown
in Fig. 9 which indicates that the cooling capacities are maximized at
the switching times around 30s, 40s, 40, and 40s for Type A, Type B,
Type C, and Type D combined ADRSs, respectively. It should be noted
that when the switching time is less than or more than 40s or less
than or more than 30s, the pressure of adsorbers will less than or
more than the pressure of evaporators. Consequently, some parts of
adsorption processes will occur at variable pressures in spite of the
pressure of adsorption processes at 40s and 30s switching times.
Because maximum adsorptions take place at the same pressures
with pressures of evaporators, the cooling capacities will be lower
than the cooling capacity at 40s and 30s switching times.
Figs. 10 and 11 indicate temperature profiles of different com-
ponents for all types of combined ADRS for both upper and bottom
cycles, respectively. Outlet temperature profiles of heat transfer
fluids for all types of combined ADRS are shown in Figs. 12 and 13
for both upper and bottom cycles, respectively. It is observable from
Fig. 12 that the cooling water outlet temperatures from condensers
of higher cycles are around 3  C, 5  C, 5  C, 4  C higher than the
Fig. 23. Sensitivity analysis of cooling capacity and COP for Type D combined ADRS.
cooling water inlet temperatures for Type A, Type B, Type C, and
Type D combined ADRS, respectively. Referring to Fig. 13, it can be
seen that chilled water outlet temperatures supplying refrigeration
capacities are about 5  C, 7  C, 5  C, 7  C lower than chilled water
inlet temperatures, for Type A, Type B, Type C, and Type D com-
bined ADRS, respectively.
The effect of different parameters on combined ADRS perfor-
mances have been evaluated by selecting maximum adsorption/
desorption time and maximum pre-heating/pre-cooling time for
each of the combined ADRS types. By considering the mentioned
times the effect of cooling water and hot water temperature has
been shown in Figs. 14e19.
According to Figs. 14e16, by increasing the cooling water inlet
temperature or by decreasing the hot water inlet temperature, the
cooling capacity will decrease in each type of combined ADRS. It is
clear from Figs. 17e19, COP decreases by increasing the cooling
water temperature in each type of combined ADRS.
According to Table 4 and Figs. 7 and 9, the performance of
different combined ADRS are presented in Table 5.
7. Sensitivity analysis
The purpose of doing the sensitivity analysis in the present
study is finding the important parameters which experimenters
should determine them in laboratories with double precision. So,
we have done sensitive analysis for different type of combined
ADRS to determine important parameters.
Equations (16)e(19) show that the cooling capacity and coeffi-
cient of performance are controlled by several parameters. To
explore the controlling parameters, a sensitivity analysis has been
performed on the presented models using Saltelli et al. [28]. A
sensitivity parameter (Si) is defined as:
   
ln y1;i  ln y0;i
Si ¼     (20)
ln x1;i  ln x0;i
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199 197

Fig. 24. More affective parameters on cooling capacity and COP of Type A combined ADRS.

Fig. 25. More affective parameters on cooling capacity and COP of Type A combined ADRS.

Fig. 26. More affective parameters on cooling capacity and COP of Type A combined ADRS.

Fig. 27. More affective parameters on cooling capacity and COP of Type A combined ADRS.
198 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199

Table 6 accompanied with non-physical results. In addition, the result of

Important parameters based on sensitivity analysis. sensitivity analysis shows that the most important performance
Type Capacity cooling COP parameters on the performance of combined ADRS are Thw, in
Parameter Sensitivity [%] Parameter Sensitivity [%]
and Tcw (ads, cond), in that their variation should be investigated on
the performance of combined ADRS in any analytical, experi-
A Thw, in 20 Thw, in 7
mental, and numerical studies.
Tcw (ads, cond), in 19 Qis, zeolite 27
hfg 17 hfg 34
B Tcw (ads, cond), in 20 hfg 24
References
Thw, in 18 Easilica gel 16
Easilica gel 15 Qis, zeolite 14
[1] R.Z. Wang, Adsorption refrigeration research in Shanghai Jiao Tong University,
C Easilica gel 26 hfg 28
Renew. Sustain. Energy Rev. 5 (2001) 1e37.
Tcw (ads, cond), in 20 Easilica gel 19
[2] U. Desideri, S. Proietti, P. Sdringola, Solar-powered cooling systems Technical
hfg 8 Tcw (ads, cond), in 13 and economic analysis on industrial refrigeration and air conditioning appli-
D Easilica gel 31 Easilica gel 33 cation, Appl. Energy 86 (2009) 1376e1386.
Tcw (ads, cond), in 21 Tcw (ads, cond), in 17 [3] B.B. Saha, E.C. Boelman, T. Kashiwagi, Computer simulation of a silica gel-
Thw, in 16 hfg 13 water adsorption refrigeration cycle e the influence of operating conditions
on cooling output and COP, ASHRAE Trans. 101 (1995) 348e357.
[4] H.T. Chua, K.C. Ng, A. Malek, T. Kashiwagi, A. Akisawa, B.B. Saha, Modeling the
performance of two-bed, silica gel-water adsorption chiller, Int. J. Refrig. 22
(1999) 194e204.
where, x0 and x1 are the initial and final values of the parameter to [5] E. Boelman, B.B. Saha, T. Kashiwagi, Experimental investigation of a silica gel-
be examined, and y0 and y1 are the corresponding model solutions water adsorption refrigeration cycle-the influence of operationg conditions on
in cooling capacity or coefficient of performance. The normalized cooling output and COP, ASHRAE Trans. 101 (1995) 358e366.
[6] M. Mahdavikhah, H. Niazmand, Effect of plate finned heat exchanger pa-
sensitivity analysis value (SiN) is calculated by dividing the calcu- rameters on the adsorption chiller performance, Appl. Therm. Eng. 50 (2013)
lated Si value by Si,max, representing the largest value calculated for 939e949.
the model. Each parameter alters by 10% in each type of combined [7] H. Niazmand, H. Talebian, M. Mahdavikhah, Effects of particle diameter on
performance improvement of adsorption systems, Appl. Therm. Eng. 59
ADRS, and the resulting sensitivity is shown in Figs. 20e23, (2013), 243e252.
respectively. [8] A. Sakoda, M. Suzuki, Fundamental study on solar powered cooling systems,
Figs. 24e27 are presented to understand the sensitivity of Chem. Eng. J. 17 (1984) 52e57.
[9] L.Z. Zhang, A three-dimensional non-equilibrium model for an intermittent
important parameters on the cooling capacity and COP precisely.
adsorption cooling system, Sol. Energy 69 (1) (2000) 27e35.
According to the Figs. 24e27, the most important parameters in [10] L.Z. Zhang, Design and testing of an automobile waste heat adsorption cooling
defining the cooling capacity and COP of each type of combined system, Appl. Therm. Eng. 20 (2000) 103e114.
[11] H.M. Lai, An enhanced adsorption cycle operated by periodic reversal forced
ADRS are presented in Table 6. Determining the correct value of
convection, Appl. Therm. Eng. 20 (2000) 595e617.
more affective parameters needs to be evaluated with more [12] Y.I. Aristov, G. Restuccia, G. Cacciola, V.N. Parmon, A family of new working
attention; otherwise, the results of performance analyses will be materials for solid sorption air conditioning systems, Appl. Therm. Eng. 22 (2)
accompanied with non-physical results. (2002) 191e204.
[13] Y.I. Aristov, New family of materials for adsorption cooling: material scientist
approach, J. Eng. Thermophyics 16 (2) (2007) 63e72.
[14] Y.I. Aristov, L.L. Vasiliev, New composite sorbents of water and ammonia for
8. Conclusion chemical and adsorption heat pumps (review), J. Eng. Thermophyics 79 (6)
(2006) 1214e1229.
[15] K.C. Chan, C.Y.H. Chao, G.N. Sze-To, K.S. Hui, Performance predictions for a
In this work, different arrangements of two adsorbents, Silica gel new zeolite 13X/CaCl2 composite adsorbent for adsorption cooling systems,
(RD type) and Zeolite 13X/CaCl2, have been used to simulate Int. J. Heat Mass Transf. 55 (2012) 3214e3224.
combined ADRS. Using the Simulink of MATLAB software, the [16] A. Sadeghlu, M. Yari, S.M.S. Mahmoudi, H. Beidaghy Dizaji, Performance
evaluation of Zeolite 13X/CaCl2 two-bed adsorption refrigeration system, Int.
analysis has been performed by applying the conservation of mass
J. Therm. Sci. 80 (2014) 76e82.
and energy equations for different system components. The [17] B.B. Saha, A. Chakraborty, S. Koyama, Y.I. Aristov, A new generation cooling
following results have been obtained: device employing CaCl2-in-silica gelewater system, Int. J. Heat Mass Transf.
52 (2009) 516e524.
[18] B.B. Saha, A. Akisawa, T. Kashiwagi, Silica gel water advanced adsorption

The cooling capacity of combined ADRS with Zeolite 13X/CaCl2- refrigeration cycle, Energy 22 (4) (1997) 437e447.
water working pair is more than the other working pair. [19] B.B. Saha, S. Koyama, K.C. Ng, Y. Hamamoto, A. Akisawa, T. Kashiwagi, Study
Therefore, whenever we seek much more cooling capacity, on a dual-mode, multi-stage, multi-bed regenerative adsorption chiller,
Renew. Energy 31 (13) (2006) 2076e2090.
Zeolite 13X/CaCl2-water working pair is preferred to Silica gel [20] F.P. Song, L.X. Gong, L.W. Wang, R.Z. Wang, Study on gradient thermal driven
(RD type)-water working pair. adsorption cycle with freezing and cooling output for food storage, Appl.

If the cooling water inlet temperature increases or the hot water Therm. Eng. 70 (2014) 231e239.
[21] K. Habib, B.B. Saha, A. Chakraborty, S. Koyama, K. Srinivasan, Performance
inlet temperature decreases, the cooling capacity for both evaluation of combined adsorption refrigeration cycles, Int. J. Refrig. 34 (2011)
working pairs will decrease. 129e137.

By increasing cooling water inlet temperature, COP of combined [22] K. Habib, B.B. Saha, A. Chakraborty, S.T. Oh, S. Koyama, Study on solar driven
combined adsorption refrigeration cycles in tropical climate, Appl. Therm.
ADRS decreases. Eng. 50 (2) (2013) 1582e1589.

The arrangement of adsorbents affects cooling capacity of [23] M. Ma, J. Yu, X. Wang, Performance evaluation and optimal configuration
combined ADRS significantly. Type A Zeolite 13x/CaCl2-water analysis of a CO2/NH3 cascade refrigeration system with falling film evapo-
ratorecondenser, Energy Convers. Manag. 79 (2014) 224e231.
has been used as an adsorbent for both top and bottom cycles, [24] R.F. Peterson, “Evaporator-condenser unit,” US patent, publication number:
not only does have more cooling capacity than the other types, US3149479 A, publication type: Grant, publication date: Sep 22, 1964, filing
but also the effect of hot water temperature on cooling capacity date: May 1, 1961.
[25] M. Polanyi, The potential theory of adsorption, Science 141 (1963)
of this type is less than the others.
1010e1013.

The result of sensitivity analysis shows that more affective [26] T. Miyazaki, A. Akisawa, B.B. Saha, I.I. El-Sharkawy, A. Chakraborty, A new
constant parameters on the performance of combined ADRS are cycle time allocation for enhancing the performance of two-bed adsorption
Qis, zeolite, Easilica gel, hfg which experimenters should determine chillers, Int. J. Refrig. 32 (2009) 846e853.
[27] C.Y. Tso, K.C. Chan, Christopher Y.H. Chao, C.L. Wu, Experimental performance
these parameters in laboratories with double precision; other- analysis on an adsorption cooling system using zeolite 13X/CaCl2 adsorbent
wise, the results of analytical and numerical studies will be with various operation sequences, Int. J. Heat Mass Transf. 85 (2015) 343e355.
 A. Sadeghlu et al. / Applied Thermal Engineering 87 (2015) 185e199 199

[28] A. Saltelli, K. Chan, E.M. Scott, Sensitivity Analysis, LTD.: John Wiley and Sons, T: temperature, K
2000. tcycle: cycle time, s
t: time, s
U: overall heat transfer coefficient, kW/m2 K
xeq: adsorption equilibrium uptake, kg/kg
Nomenclature x: instantaneous adsorption uptake, kg/kg

subscript
Symbol
a: adsorbent
A: area, m2 ads: adsorption
COP: coefficient of performance, [] bed: sorption element (adsorber/desorber)
cp: specific heat capacity, kJ/kg K chill: chilled water
Ds: surface diffusivity, m2/s cond: condenser
Dso: pre-exponential term, m2/s cw: cooling water
Ea: activation energy, kj/kmol des: desorption
h: enthalpy, kJ/kg evap: evaporator
Ksav: mass transfer coefficient, 1/s fHex: fin (aluminum)
M: mass, kg hw: hot water
_ mass flow rate, kg/s
m: in: inlet condition
Qis: isosteric heat of adsorption, kJ/kg kHex: heat transfer tube (copper)
Q: power, kW out: outlet condition
R: universal gas constant, kJ/kmol K ref: refrigerant (adsorbate)
Rp: absorbent particle radius, m w: water
SCP: special cooling power, kW/kg