Energy 36 (2011) 530e537

Contents lists available at ScienceDirect

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

Simulation of an adsorption solar cooling system

H.Z. Hassan a, A.A. Mohamad a, *, R. Bennacer b
a
Mechanical and Manufacturing Engineering, Schulich School of Engineering, University of Calgary, 2500 univ. Drive, NW, Calgary, Alberta, Canada T2N 1N4
b
University de Cergy-Pontoise, L2MGC F-95000 Cergy-Pontoise, France

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

Article history: A more realistic theoretical simulation model for a tubular solar adsorption refrigerating system using
Received 6 February 2010 activated carbonemethanol (AC/M) pair has been introduced. The mathematical model represents the
Received in revised form heat and mass transfer inside the adsorption bed, the condenser, and the evaporator. The simulation
3 August 2010
technique takes into account the variations of ambient temperature and solar radiation along the day.
Accepted 6 October 2010
Available online 3 November 2010
Furthermore, the local pressure, and local thermal conductivity variations in space and time inside the
tubular reactor are investigated as well. A Cþþ computer program is written to solve the proposed
numerical model using the finite difference method. The developed program covers the operations of all
Keywords:
Solar energy
the system components along the cycle time. The performance of the tubular reactor, the condenser, and
Adsorption the evaporator has been discussed. Time allocation chart and switching operations for the solar refrig-
Refrigeration eration system processes are illustrated as well. The case studied has a 1 m2 surface area solar flat plate
Modelling collector integrated with a 20 stainless steel tubes containing the AC/M pair and each tube has a 5 cm
Simulation outer diameter. In addition, the condenser pressure is set to 54.2 kpa. It has been found that, the solar
Activated carbonemethanol coefficient of performance and the specific cooling power of the system are 0.211 and 2.326 respectively.
In addition, the pressure distribution inside the adsorption bed has been found nearly uniform and
varying only with time. Furthermore, the AC/M thermal conductivity is shown to be constant in both
space and time.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction refrigeration systems with environmentally friendly ones that can

Solar radiation is by far the largest and the most world’s abun-
dant, clean and permanent energy source. The amount of solar
radiation intercepted by the Earth’s surface is much higher than the
annual global energy use. The energy available from the sun
ð821015 WÞ is greater than about 5200 times the global world’s
need in 2006. In recent years, many promising technologies have
been developed to harness the Sun’s energy. These technologies help
in environmental protection, economizing energy, and sustainable
developments which are the major issues of the world in the 21st
century.
One of these important technologies is the solar cooling system
that makes use of either absorption or adsorption technologies. Due
to the environmental problems of chlorofluorocarbons (CFCs)
emissions, adsorption cooling technologies are more attractive field
of research and development than the conventional vapour
compression refrigeration systems. The international and local
policies nowadays are directed towards replacing the traditional
* Corresponding author. Tel.: þ1 403 220 2781; fax: þ1 403 282 8406.
E-mail address: mohamad@ucalgary.ca (A.A. Mohamad).
be operated by new and renewable energy sources [1]. The solar
adsorption cooling systems are good alternative since they operate
with environmentally benign refrigerants that are natural, free
from CFCs and therefore they have a zero ozone depleting potential
(ODP). Furthermore, these refrigerants satisfy the Kyoto protocol on
global warming, the Vienna Convention for Protection of the Ozone
Layer (1985), and Montreal Protocol on Substances Depleting the
Ozone Layer (1987).
Adsorption cooling systems are characterized by their simple
control and the absence of vibration and corrosion problems. The
wide range of heat source temperatures that can be used
50  C 600  C, and the low operation and maintenances costs make
these systems more attractive [2]. Although the adsorption chiller
systems have these advantages, their drawbacks are the intermit-
tent operation, the requirements of special designs to maintain
high vacuum, the large volume and weight relative to traditional
refrigeration systems, the low specific cooling power (SCP) and the
low coefficient of performance (COP) [3]. Besides the poor heat and
mass transfer within the adsorbent, the adsorption deterioration of
the adsorbent is also vital to the development and applications of
the adsorption refrigeration technology [4]. However, enhance-
ment of heat and mass transfer properties in the adsorbent bed,

0360-5442/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2010.10.011
 H.Z. Hassan et al. / Energy 36 (2011) 530e537 531

Nomenclature Upoepi Overall heat transfer coefficient between the outer and
inner surfaces of the pipe based on the pipe inner
A Area [m2] surface area [W m2 K1]
C Specific heat [J kg1 K1] V Volume [m3]
hceamb Convection heat transfer coefficient between the
condenser and the ambient [Wm2 K1] Greek letters
Imax The maximum solar radiation intensity [W m2] d Thickness [m]
L Length [m] DH Isosteric heat of adsorption [Jkg1].
Wo Constant in the DubinineAstakhov (DeA) equation 3 Porosity of the medium [-]
(Eq. (6)) [m3 kg1] r Density, kgm3
x Adsorbate concentration ratio [kg/kg] s Stefan Boltzmann constant [W m2 K4]
Cp Specific heat at constant pressure [J kg1 K1] s Transmissivity [-]
Cv Specific heat at constant volume [J kg1 K1] 3 Emissivity [-]
D Constant in the DubinineAstakhov (DeA) equation a Absorptivity
(Eq. (6)), or diameter [m]. q Volume fraction of the adsorbed phase [-]
hgoeamb Convection heat transfer coefficient between the outer
glass cover and the ambient [W m2 K1] Subscripts
k Coefficient of thermal conductivity [W m1 K1] a Adsorbed phase.
Lv Latent heat of vaporization [J kg1] amb Ambient
M Total mass [kg] cond Condensation
m Mass [kg] d Desorbed
mcl Mass of liquid methanol that leaves the condenser eff Effective
towards the evaporator [kg] evap Evaporation or evaporator
n Constant in the DubinineAstakhov (DeA) equation g Glass cover or the gas phase
(Eq. (6)) l Liquid
P Pressure [pa] o Out
Qe Refrigeration effect [J] p Pipe
R Radius [m] pi Inner surface of the pipe
T Temperature [K] s Solid phase or saturation
t Time [sec] po Outer surface of the pipe
tsr The time of the day when sun rises [sec]
tss The time of the day when sun sets [sec] Abbreviations
U Overall heat transfer coefficient [W m2 K1] AC/M Activated carbonemethanol
Ugoepo Overall heat transfer coefficient between the outer CFCs Chlorofluorocarbons
surface of the glass cover and the outer pipe surface COP Coefficient of performance
based on the glass cover surface area [W m2 K1] ODP Ozone depleting potential
Upoeamb Overall heat transfer coefficient between the outer SCOP Solar coefficient of performance
surface of the pipe and the ambient based on the glass SCP Specific cooling power [Wkg1]
cover surface area [W m2 K1]

increasing the adsorption properties of the working pairs and
a better heat management during the adsorption cycle lead to
a more efficient system [2].
The performance of the adsorption cooling system depends
mainly on the working pairs used. A good designed system should
have the characteristics of large adsorption capacity, large change of
adsorption capacity with temperature variation, and a more flat
desorption isotherm. Moreover, the refrigerant should have a large
latent heat per unit volume, no toxicity, no flammability, no
corruption, and good chemical and thermal stability [4]. The most
widely used working pairs are activated carbonemethanol, activated
carbon fiberemethanol, activated carboneammonia, zeoliteewater,
silica gelewater, calcium chlorideeammonia and composite adsor-
benteammonia pairs.
Anyanwu [3] presented a review of the practically realized solid
adsorption solar refrigeration cycles. He classified the cycles accord-
ing to the adsorbate utilized as: cycles with water as refrigerant,
cycles using fluorocarbon as refrigerant, cycles using ammonia as
refrigerant and cycles with alcohols as refrigerant. Activated-carbon
is the most widely used adsorbent reported in literature due to its
extremely high surface area and micro pore volume. Moreover, Cri-
toph [5] and Critoph and Vogel [6] found charcoal a preferable and
recommended adsorbent for solar cooling. Activated carbon-
emethanol pair has been studied by many researchers [7e9]. This
pair is characterized by high latent heat of vaporization of methanol
and low desorption temperature. However, the operating tempera-
ture is one of the limitation that constrain the operation of the acti-
vated carbonemethanol pair cooling systems. At temperatures more
than 150  C, the methanol decomposes into dimethyl. However,
a very low decomposition reaction rate is found to happen at
temperatures near 120  C, Yuan [10]. The simulation done by Eric [11]
showed that the thermal decomposition of methanol did occur in the
normal solar powered adsorption refrigeration system. The rate of the
decomposition depends on the presence of various alloys such as
aluminium and copper alloys. Moreover, the decomposition products
may be one of the main reasons for the diminishing performance of
the solar ice maker.
The solar adsorption refrigeration system usually operates with
the basic simple cycle. The basic cycle is a single-bed, intermittent,
and has a low efficiency and low cooling effect. Many attempts have
been reported in the literature to develop and design a semi-
continuous operation and to increase the performance and efficiency
of the cycle. These developments include heat recovery or the
regeneration cycle [12e15], mass recovery cycle [16e18], heat and
532 H.Z. Hassan et al. / Energy 36 (2011) 530e537

mass recovery cycle [13,19], thermal wave cycle [20], forced convec-
tive thermal wave cycle [21e23], cascade cycle [24e29], multi-bed
cycle [30e32], multi-stage cycle [33e37], dual-mode cycles [38]. An
explained review for these cycles is presented in [3,39].

2. Physical model

The adsorption cooling system likes the basic vapour compression

refrigeration machine except that the power compressor is replaced
with a thermal compressor or a reactor. This reactor is composed of
a type of porous medium that has the ability to adsorb the refrigerant.
The working principle of the basic reactor cycle is represented in the
Clapeyron diagram, Fig. 1. The adsorbenteadsorbate cycle consists
mainly of four phases; pressurization process at a constant volume
(isosteric heating phase), desorption at constant pressure (isobaric
heating phase), depressurization at constant volume (isosteric cool-
ing phase), and adsorption at constant pressure (isobaric cooling
phase).
At the beginning of the day, state A, the reactor is isolated from
both the condenser and the evaporator by valves c and e and is
completely charged with the refrigerant. The pressure inside the
reactor initially equals the evaporator pressure Pevap. and its
temperature is uniform and equals the ambient temperature Tamb.
When the reactor starts to heat up by the incident solar radiation,
both pressure and temperature inside the adsorbent are elevated.
This constant concentration heating phase continues till point B
where the pressure reaches a value that equals the condenser
pressure Pcond. This period is equivalent to the compression in the
classical vapour compression refrigeration cycle. At state B, valve c
is opened and the adsorbate starts to desorb and flow towards the
condenser. During this isobaric heating phase, the temperature
continues to increase, and the adsorbate concentration continues to
decrease as more adsorbate is being freed from the reactor. When
the adsorbate temperature reaches the maximum value Tmax at
state C, valve c is closed and the reactor starts the third phase.
When the solar flux decreases, the reactor is cooled down at
constant volume and constant isoster till the pressure inside the
reactor decreases to the evaporator pressure Pevap. , point D. The last
phase of the reactor cycle starts at the night, point D, when valve e
is opened and the refrigerant flows towards the reactor. The
adsorption process continues while the reactor is cooled at the
constant evaporator pressure till the higher cycle isoster at point A.
Fig. 2. Configuration diagram of the solar adsorption cooling reactor.
The solar refrigeration system presented in this study has three
main components: the solar collector/reactor, Fig. 2, the condenser,
and the evaporator. The solar collector/reactor is constituted of
a clear plane glass sheet cover, and the activated carbonemethanol
pair which is contained in the annular space between a two coaxial
collector pipes. The inner pipe is perforated in order to ease the
flow of methanol into and out from the activated carbon granules
and to avoid pressure drops and temperature differences along the
collector tubes as well. These pipes are integrated in the plane solar
collector and are coated with a selective painting to allow a good
absorption of solar radiation and low emission. The lateral and rear
sides of the collector are insulated. The inner pipes are connected to
a common methanol inlet and outlet headers. The adsorbent bed is
cooled during the adsorption process when the collector glass
cover plate is opened by natural convection and by radiation from
the collector plate and tubes. The condenser pipe is steel and is
cooled by natural convection and by radiation as well.
3. Mathematical model
The presented mathematical model simulates the real working
of the solar adsorption cooling system taking in account the vari-
ation of solar radiation and ambient temperature along the day.
Furthermore, the energy and mass balances for every element of
the system are taken into account as well.
3.1. Solar radiation and ambient temperature modelling
Both desorption and isosteric heating phases depend strongly on
the quantity of incident solar radiation. In addition, the adsorption

Fig. 1. Clapeyron diagram for a conventional adsorption cycle.

 H.Z. Hassan et al. / Energy 36 (2011) 530e537 533

and the isosteric cooling phases besides the condenser operation are lower temperature layers. This mass transfer results in the heat
dependent on the ambient temperature. Therefore, in order to transfer process. That is because the gas that will be adsorbed on
simulate the system in a more realistic manner, the solar radiation cold layers is hot. Therefore, the adsorption and desorption mass
and ambient temperature are modelled in this study as variant along transfer through the porous medium is accompanied with a heat
the day. These climatic conditions are taken for Calgary town, transfer and vice versa. The conservation of mass equation and the
Canada (latitude ¼ 51.12 and longitude ¼ 114.01) where a summer first law of thermodynamics for an open system are applied to
day in July 30, 2009 is chosen. The apparent sun rise time, apparent a layer of radial coordinates r and thickness dr in the adsorbent bed.
sun set time, and the solar noon during this day are at 5:59:00, Both the mass and heat balance equations are merged together to
21:25:00, and 13:42:30 respectively [43]. The intensity of solar give the following combined heat and mass transfer equation
radiation as a function of the time is given by: within the adsorbent bed:


 
pðt  tsr Þ P vT
Is ðtÞ ¼ Imax sin (1) ð1  3Þrs Cs þ qra Ca ð3  qÞ rg Cvg þ
tss  tsr T vt
" # 
2
where, tsr < t< tss v T 1 vT vP P vra vq
¼ Keff þ þ ð3  qÞ þ q þ DH þ ra DH
Imax is taken to be equal 900 W/m2. vr 2 r vr vt ra vt vt
Furthermore, the real hourly temperature variation on the same
(4)
day is taken from [44]. The temperature data is fitted in a form of
a polynomial from the 8th degree and is inserted to the simulation In this equation, the adsorbate density is modelled as varying
program. with temperature. Moreover, the pressure is taken as variant
throughout the adsorbent bed layers. It is assumed that the gaseous
3.2. Energy balance for the solar collector phase behaves as an ideal gas. The effective thermal conductivity of
the adsorbent bed is also taken as variant and is evaluated by the
The solar radiation is absorbed at the external surface of every following formula:
tube. This heat flux is transmitted by conduction, through the pipe
Keff ¼ Ks ð1  3Þ þ Ka q þ Kg ð3  qÞ (5)
wall in the inward radial direction to the internal walleadsorbent
interface, then towards the adsorbent-refrigerant reactive medium. where ks,ka, and kg are the thermal conductivities of the solid,
A one-dimensional transient model for heat balance is suggested adsorbate, and gas phases respectively.
for the glass cover and the pipe. The heat balance for the glass cover
is given by the following equation:

vTgo   3.4. Adsorption equilibrium model

rg Cg dg Ag ¼ ag Ag Is ðtÞ þ Ugopo Ag Tpo  Tgo
vt
 
4 4 The adsorption equilibrium equation of the activated carbon/
  s Tpo  Tgo pRpi Lp
 hgoamb Ag Tgo  Tamb þ  methanol as an adsorbent/adsorbate pair presented in this study, is
1 1
þ 1 described by the DubinineAstakhov (DeA) equation of state. The
3p 3g adsorbed mass of methanol per unit mass of activated carbon is
 
4  T4
s Tgo given by:
amb

  n
1 1 Ps ðTÞ
þ 1 X ¼ ra ðTÞWo Exp  D Tln (6)
3g 3amb P
(2)
The values of the constants Wo, D, and n are given in Table 1.
Neglecting the heat conduction between the adsorbent bed and The thermodynamic properties of methanol are taken from
the condenser and evaporator, the energy balance for the steel pipe Methanex Corporation [40]. In order to complete the mathematical
is given by: formulation of the model, the methanol properties are fitted as
polynomials and merged with the mathematical model.
vTpo  
rp Cp Vg ¼ sg ap Is ðtÞLp Dpo  Ugopo Ag Tpo  Tgo
vt
 
4 4
  s Tpo  Tgo pRpi Lp
 Upopi pDpi Lp Tpo  Tpi   3.5. Condenser, evaporator and system performance
1 1
þ  1 Ag
3p 3g The mass of methanol gas desorbed from the adsorption bed
 
4 goes to the condenser where it is cooled and condenses to liquid.
  s Tpo  Tg4 pRpi Lp
 Upoamb Ag Tpo  Tamb   The condenser is cooled by the convection and radiation to the
1 1 ambient air. The combined mass and energy conservation equation
þ  1 Ag
3g 3p of the condenser is given by:
(3)
Table 1
Constants of the DubinineAstakhov (DeA) equation, Jing and Exell[7].

3.3. Combined heat and mass transfer equation in the adsorbent Charcoal Wo[Lit/kg] D  105 n
bed 207E4 0.36546 14.96202 1.34
Chinese LSZ30 0.405 31.972 1.26
LH 0.860 25.74 1.321
It is clear that, higher temperature layers in the adsorbent bed Thai MD6070 0.988 88.98 1.12
desorb the methanol gas. This desorbed gas is adsorbed on the
534 H.Z. Hassan et al. / Energy 36 (2011) 530e537

vTc   simulate the behaviour of the solar adsorption cooling system. For
½mc Cc þ Md ðtÞCd  ¼ md ðtÞ Lv ðPc Þ þ Cpg Tg  Tcond each time step, the program updates the ambient temperature and
vt
the solar insolation based on the suggested climatic scheme. Based
 Ac hcamb ðTcond  Tamb Þ
  on the new conditions every time step, the temperature, pressure,
s Tc4  Tamb4 concentration ratio, volume fraction, desorbed and adsorbed mass
 (7) inside the reactor, and the condenser temperature are updated. At
1 1
þ  1 Ag the end of the cycle, which is adjusted to be within 24 h period, the
3c 3amb
overall system performance parameters are determined. These
The cooling effect is calculated by the following equation: parameters include the solar coefficient of performance, the
      specific cooling power, and the evaporation effect. The main data
Qe ¼ mcl Lv Tevap  hl Tc;o þ hl Tevap (8) and design parameters input to the developed Cþþ code and used
in this simulation are depicted in Table 2.
The parameters used in this present work to assess the
adsorption refrigeration system performance are the solar coeffi-
cient of performance SCOP and the specific cooling power SCP. 5. Results and discussion
These parameters are defined by the following equations:
Based on the proposed mathematical and numerical models and
Qe the input data presented in Table 2, the output results for only one
SCOP ¼ Z (9)
cycle of operation are presented in this section. The obtained
Is ðtÞAg dt
results and the overall system performance are estimated in this
study by considering a number of 20 pipes backed with the acti-
The SCP is the ratio between the cooling effect and the cycle
vated carbonemethanol pair and placed in the tubular reactor. Each
time per unit mass of adsorbent:
tube is 1 m in length and the surface area of the solar flat plate
Qe
SCP ¼ (10)
tcycle ms Table 2
The main parameters used in the case study.
where tcycle is the complete cycle time.
Symbol Parameter Value Unit
Glass cover
3.6. Initial and boundary conditions Cg Specific heat of the glass cover 750 [J kg1 K1]
rg Density of the glass cover 2500 [kg m3]
The operation of the solar cooling system is strongly dependant tg Glass cover thickness 0.003 [m]
kg Coefficient of thermal conductivity 1.4 [W m1 K1]
on the variation of the climatic conditions. This means that, the
sg Transmissivity 0.95 []
performance of this system changes day by day. Moreover, the ag Absorptivity 0.05 []
initial conditions of the system for a given cycle are defined by the 3g Emissivity 0.9 []
final conditions for the previous one. In this work we deal with the Air gap
first operating cycle in which the system is considered in thermal tgab Air gap thickness 0.1 [m]
kair Air thermal conductivity 0.0263 [W m1 K1]
equilibrium with the ambient. Therefore, the initial pressure and
Back insulation
temperature are given by: tins Insulation thickness 0.1 [m]
  kair Insulating material thermal conductivity 0.038 [W m1 K1]
Pðt ¼ 0Þ ¼ Pevap ¼ Psat Tevap Stainless steel pipe
Cp Specific heat of stainless steel 480 [J kg1 K1]
rp Density of stainless steel 8055 [k gm3]
Tðt ¼ 0Þ ¼ Tgo ðt ¼ 0Þ ¼ Tpo ðt ¼ 0Þ ¼ Tpi ðt ¼ 0Þ kp Coefficient of thermal conductivity 15.1 [W m1 K1]
ap Absorptivity of the pipe coating material 0.98 []
¼ Tc ðt ¼ 0Þ ¼ Tamb 3p Emissivity of the pipe coating material 0.1 []
Dpo Outer diameter of the pipe 0.05 [m]
The boundary conditions used to solve equation (4) are as Dpi Inner diameter of the pipe 0.048 [m]
follows : Lp Pipe length 1.0 [m]
np Number of pipes 20 []
vTpi   Adsorption bed
1: At r ¼ Rpi ; Keff ¼ Upopi Tpo  Tpi
vr 3 Porosity of the AC 0.7 []
Cs Specific heat of carbon 711 [J kg1 K1]
ks Thermal conductivity of carbon 1.6 [W m1 K1]
vT rs Density of carbon 2000 [k gm3]
2: AT r ¼ Ri ; ¼ 0:0
vr Dip Inner diameter of the inner coaxial pipe 0.01 [m]
Methanol properties
The second boundary condition assumes an adiabatic condition Ca Specific heat of liquid methanol 2534 [J kg1 K1]
at the inner tubular reactor surfaceegas interface [41,42]. This can Cpgas Specific heat of methanol vapour atp ¼ cons: 1820 [J kg1 K1]
be explained by the poor thermal conductivity of methanol gas Cvgas Specific heat of methanol vapour at v ¼ cons: 1560 [J kg1 K1]
ka Thermal conductivity of liquid methanol 0.2022 [W m1 K1]
which is about 0.016 W/m K.
kgas Thermal conductivity of methanol vapour 0.0163 [W m1 K1]
Condenser
Pcond Condenser Pressure 54,209 [Pa]
4. Numerical method of solution
Lc Condenser tube length 20 [m]
Roct Condenser tube outer radius 0.0025 [m]
In the present work, the proposed mathematical model is dis- Rict Condenser tube inner radius 0.002 [m]
cretized using finite difference approximation method. The Ccm Specific heat of the condenser metal 480 [J kg1 K1]
resulting system of equations is solved explicitly to obtain the rcm Density of the condenser metal 8055 [k gm3]
Evaporator
numerical solution for the variables at every time step. A computer Pevap Evaporator Pressure 3967 [Pa]
code based on this numerical scheme is written in Cþþ in order to
 H.Z. Hassan et al. / Energy 36 (2011) 530e537
Fig. 3. Variation of climatic data with the day time.
collector glass cover is 1 m2. Moreover, the condenser pressure is
set to a value of 54.21 kpa which corresponds to a saturation
temperature of 50  C. Generally, it is found that the solar coefficient
of performance of this adsorption cooling system attains a value of
0.211. Furthermore, the specific cooling power and the evaporator
cooling effect values are found to be 2.326, and 4175 kJ respectively.
Fig. 3 shows the variation of the solar incident radiation and the
ambient atmospheric temperature with the day hours. These data
simulate a summer day in July 30, 2009, for Calgary, Canada as
discussed in Section 3.1. The time allocation chart for the solar
refrigeration system is illustrated in Fig. 4. This chart indicates the
rank of each process, its period of operation, and its start and end
time. The timing schedule is affected strongly by the climatic
conditions, cycle initial conditions, and the system design param-
eters. Moreover, the behaviour of any phase of the reactor affects
the operation of the following other phases and therefore the
condenser, evaporator, and the overall cycle of the system. There-
fore, the matching among the system elements is an important
issue that insures a proper operation of the cooling system and
satisfies a daily complete working cycle.
Fig. 5 represents the evolution of the volume averaged
temperature of the adsorbent bed along the cycle period. The
period of the cycle is 24 h starting at 6:00 in the morning which is
the apparent sun rise time. It is noticed that, during the isosteric
heating phase, the reactor continues to absorb the continuously
increasing solar radiation. This results in a rapid increase in the
adsorption bed temperature from about 10.8  C to 60  C in a time
Fig. 4. The time allocation of the solar cooling system.
535
Fig. 5. Variation of average adsorption bed temperature with time.
period of about 3.55 h. The pressure inside the activated carbon-
emethanol pair approaches the condenser pressure of 54.209 kpa
at the temperature of 60  C. At this moment, methanol starts to
desorb from the adsorbent and flows towards the condenser.
During this process, heat energy is extracted from the adsorption
bed due to the endothermic character of the desorption process.
Therefore, the rate of heating and the associated temperature
increase rate during this isobaric desorption phase is lower than the
first phase as seen from Fig. 5. The desorption process continues for
a time period of 5.4 h and stops when methanol reaches 120  C, the
temperature at which it starts to decompose. At this moment, the
solar flat plate collector glass cover is opened and the reactor is kept
away from the sun rays to allow beginning of the cooling phase of
the adsorption bed. Heat transfers from the reactor by means of
both natural convection and by radiation with a high driving
potential from the high reactor temperature to the low temperature
of the surroundings. This results in the sudden dramatic decreasing
rate in the reactor temperature, as noticed in Fig. 5, from 120  C to
61  C in a short time period of 0.74 h. When the temperature rea-
ches 61  C, the corresponding reactor pressure at this low isosteric
process will be the same as the evaporator pressure and adsorption
process has to begin. Since the adsorption process is exothermic
and the driving potential for heat transfer is lower than the
previous phase, a lower decreasing rate for the reactor temperature
is remarked in Fig. 5. The cooling of the adsorption bed continues
till it reaches a temperature of 16.3  C at the end of the cycle. It is
clear that, this temperature is not the same as the corresponding
value at the beginning of the cycle. This is because the reactor has
not reached a complete thermal equilibrium with the ambient.
Therefore, this will affect the next operating cycle of the system
which has an initial temperature of 16.3  C.
The variation of both the average pressure and the adsorbed
mass inside the activated carbonemethanol pair with time is
illustrated in Fig. 6. The average pressure increases from the initial
evaporator pressure, 3.967 kpa, to the condenser pressure,
54.209 kpa during the first phase. It reaches a plateau while the
desorption process takes place then sharply drops again to the
evaporator pressure at the beginning of the adsorption process.
Although the pressure computation in this study is modelled as
variant with the radial direction inside the tubular reactor, its
observed gradients are found to be very small and therefore can be
neglected. Consequently, it is a good assumption to consider
a uniform pressure distribution inside the adsorption bed. The
amount of adsorbed methanol at the beginning of the operating
cycle is 5.1 kg and it reaches an amount of 1:13 kg at the end of
desorption process. That is, the total effective mass of methanol
freed from the tubular reactor and circulates in the condenser and
536 H.Z. Hassan et al. / Energy 36 (2011) 530e537
Fig. 6. Variation of the average pressure and the adsorbed mass inside the adsorption
bed.
the evaporator is about 3.97 kg. Consequently, the effective meth-
anol mass ratio, which is the fraction of the adsorbate total initial
mass that circulates in the cycle, is about 0.778. During the
adsorption process, the methanol mass inside the reactor continues
to increase till it reaches a value of 4.5 kg at the end of this cycle and
beginning of the next cycle, Fig. 6.
The local values of the thermal conductivity inside the adsor-
bent bed layers are determined each time step. It is found that, the
local thermal conductivity does not vary so much from a layer to
a layer and therefore it can be considered uniform in the adsorption
bed. The change of the adsorbent bed volume averaged effective
thermal conductivity along the cycle time is seen in Fig. 7. Gener-
ally, the average conduction coefficient attains a low value that
ranges from a maximum of about 0.528 W/mK to a minimum of
about 0.5 W/mK. Since the range of variation is very small, it can be
verified that the adsorbent bed thermal conductivity is both
uniform and does not vary with time.
Fig. 8 illustrates the condenser temperature changes during the
day. It is clear that, before the adsorption process, the condenser
temperature change follows that of the corresponding ambient
temperature. As the vapour of methanol starts to flow towards the
condenser, a remarkable sudden temperature increasing rate is
seen, Fig. 8. This increase is due to the increasing desorbed methanol
vapour mass that has a high temperature. The condenser reaches
a maximum temperature of about 40  C after which its temperature
begins to go down. The condenser temperature decreases in a rapid
rate during the third phase. That is because the flow of methanol
vapour is stopped from one side and the ambient temperature starts
Fig. 7. Change of the effective thermal conductivity of the adsorbent bed along the
cycle.
Fig. 8. Condenser temperature variation with time.
to decrease providing a higher driving potential for the condenser
cooling process from the other side. This cooling process continues
by means of both free convection and radiation until the liquid
methanol reaches about 33  C at the beginning of the adsorption
process. The liquid methanol exits from the condenser to enter the
evaporator after it is throttled in the expansion valve to the evapo-
rator pressure. As the condenser begins to discharge its content of
methanol, its temperature starts to follow the ambient temperature
changes along the remaining period of the day.
6. Conclusion
In this study we have introduced a more actual theoretical
model to simulate the adsorption cooling system powered by the
solar energy. The system uses the activated carbonemethanol as
the working pair. The mathematical and numerical models for the
adsorbent bed, condenser, and evaporator has been programmed
be the Cþþ language. The real ambient temperature and solar
radiation variations are taken into account. The variation of the
adsorbent bed average temperature, pressure, adsorbed mass, and
its thermal conductivity with time are discussed. Moreover, the
condenser temperature variation, the time allocation and switching
operations and the overall cooling system performance parameters
are investigated as well. For the case studied, it is found that, the
solar coefficient of performance reaches a value of 0.211. Moreover,
the specific cooling power is estimated to about 2.326. It can also be
seen from this work that, the pressure is nearly uniform inside the
adsorption bed and the activated carbonemethanol thermal
conductivity is found to be constant in space and time.
The authors would like to acknowledge the support of Solar
Buildings Research Network and NSERC.
Acknowledgement
The authors would like to acknowledge the support of Solar
Buildings Research Network and NSERC.
References
[1] Anyanwu EE. Environmental pollution: restructuring the refrigeration
industry as a way out. Environment Protection Engineering 2000;26
(4):17e28.
[2] Wang RZ, Oliveira RG. Adsorption refrigeration e an efficient way to make
good use of waste heat and solar energy. Progress in Energy and Combustion
Science 2006;32(4):424e58.
[3] Anyanwu EE. Review of solid adsorption solar refrigerator I: an overview of
the refrigeration cycle. Energy Conversion and Management 2003;44
(2):301e12.
 H.Z. Hassan et al. / Energy 36 (2011) 530e537
[4] Wang DC, Li YH, Li D, Xi YZ, Zhang JP. A review on adsorption refrigeration
technology and adsorption deterioration in physical adsorption systems.
Renewable and Sustainable Energy Reviews 2010;14(1):344e53.
[5] Critoph RE. Performance limitations of adsorption cycles for solar cooling.
Solar Energy 1988;41(1):21e31.
[6] Critoph RE, Vogel R. Possible adsorption pairs for use in solar cooling. Inter-
national Journal of Ambient Energy 1986;7(4):183e90.
[7] Jing H, Exell RHB. Simulation and sensitivity analysis of an intermittent solar-
powered charcoal/methanol refrigerator. Renewable Energy 1994;4(1):133e49.
[8] Ogueke NV, Anyanwu EE. Design improvements for a collector/generator/
adsorber of a solid adsorption solar refrigerator. Renewable Energy
2008;33:2428e40.
[9] Leite APF, Daguenet MD. Performance of a new solid adsorption ice maker
with solar energy regeneration. Energy Conversion and Management 2000;41
(15):1625e47.
[10] Yuan ZH. Charcoal methanol adsorption refrigeration. Bangkok, Asian Insti-
tute of Technology 1988, M. Eng. Thesis ET-88e10.
[11] Hu EJ. A study of thermal decomposition of methanol in solar powered
adsorption refrigeration machines. Solar Energy 1998;62(5):325e9.
[12] Vasiliev LL, Mishkinis DA, Antukh AA, Vasiliev Jr LL. Solar-gas solid sorption
heat pump. Applied Thermal Engineering 2001;21(5):573e83.
[13] Wang RZ. Performance improvement of adsorption cooling by heat and mass
recovery operation. International Journal of Refrigeration 2001;24(7):602e11.
[14] Critoph RE. Simulation of a continuous multiple-bed regenerative adsorption
cycle. International Journal of Refrigeration 2001;24(5):428e37.
[15] Chahbani MH, Labidi J, Paris J. Modeling of adsorption heat pumps with heat
regeneration. Applied Thermal Engineering 2004;24(2e3):431e47.
[16] Alam KCA, Akahira A, Hamamoto Y, Akisawa A, Kashiwagi T. A four-bed mass
recovery adsorption refrigeration cycle driven by low temperature waste/
renewable heat source. Renewable Energy 2004;29(9):1461e75.
[17] Akahira A, Alam KCA, Hamamoto Y, Akisawa A, Kashiwagi T. Mass recovery
adsorption refrigeration cycle-improving cooling capacity. International
Journal of Refrigeration 2004;27:225e34.
[18] Luo HL, Dai YJ, Wang RZ, Wu JY, Xu YX, Shen JM. Experimental investigation of
a solar adsorption chiller used for grain depot cooling. Applied Thermal
Engineering 2006;26(11e12):1218e25.
[19] Wang W, Qu TF, Wang RZ. Influence of degree of mass recovery and heat
regeneration on adsorption refrigeration cycles. Energy Conversion and
Management 2002;43(5):733e41.
[20] Sward BK, LeVan MD, Francis M. Adsorption heat pump modeling: the
thermal wave process with local equilibrium. Applied Thermal Engineering
2000;20(8):759e80.
[21] Critoph RE. Forced convection enhancement of adsorption cycles. Heat
Recovery Systems & CHP 1994;14(4):343e50.
[22] Critoph RE. Forced convection adsorption cycle with packed bed heat
regeneration. International Journal of Refrigeration 1999;22:38e46.
[23] Hai-Ming Lai. An enhanced adsorption cycle operated by periodic reversal
forced convection. Applied Thermal Engineering 2000;20:595e617.
[24] Douss N, Meuiner FE, Sun LM. Predictive model and experimental results for
a two absorber solid adsorption heat pump. Industrial & Engineering Chem-
istry Research 1988;27(2):310e6.
[25] Douss N, Meunier F. Experimental study of cascading adsorption cycles.
Chemical Engineering Science 1989;44(2):225e35.
537
[26] Douss N. Experimental study of adsorption heat pump cycles. International
Chemical Engineering 1993;33(2):207e14.
[27] Chua HT, Ng KC, Malek A, Kashiwagi T, Akisawa A, Saha BB. Modeling the
performance of two-bed, sillica gel-water adsorption chillers. International
Journal of Refrigeration 1999;22(3):194e204.
[28] Khan MZI, Alam KCA, Saha BB, Hamamoto Y, Akisawa A, Kashiwagi T. Para-
metric study of a two-stage adsorption chiller using re-heat-The effect of
overall thermal conductance and adsorbent mass on system performance.
International Journal of Thermal Sciences 2006;45(5):511e9.
[29] Liu Y, Leong KC. Numerical study of a novel cascading adsorption cycle.
International Journal of Refrigeration 2006;29(2):250e9.
[30] Chua HT, Ng KC, Malek A, Kashiwagi T, Akisawa A, Saha BB. Multi-bed
regenerative adsorption chiller-improving the utilization of waste heat and
reducing the chilled water outlet temperature fluctuation. International
Journal of Refrigeration 2001;24(2):124e36.
[31] Saha BB, Koyama S, Lee JB, Kumahara K, Alamc KCA, Hamamoto Y, et al.
Performance evaluation of a low-temperature waste heat driven multi-bed
adsorption chiller. International Journal of Multiphase Flow 2003;29
(8):1249e63.
[32] Qenawy AM, Mohamad AA. Simulation of double-bed cooling and heating
hybrid solar adsorption refrigeration cycle. In: 2nd Canadian solar buildings
conference, Calgary; 2007.
[33] Bidyut BS, Elisa CB, Takao K. Computational analysis of an advanced adsorp-
tion-refrigeration cycle. Energy 1995;20(10):983e94.
[34] Saha BB, Boelman EC, Kashiwagi T. Computational analysis of an advanced
adsorption refrigeration cycle. Energy 1995;20:983e94.
[35] Saha BB, Kashiwagi T. Experimental investigation of an advanced adsorption
refrigeration cycle. ASHRAE Transactions 1997;103(2):50e8.
[36] Saha BB, Akisawa A, Kashiwagi T. Solar/waste heat driven two-stage adsorp-
tion chiller: the prototype. Renewable Energy 2001;23(1):93e101.
[37] Kashiwagi T, Saha BB. Experimental investigation of an advanced adsorption
refrigeration cycle. ASHRAE Transaction Research; 1997:51e8.
[38] Saha BB, Koyama S, Kashiwagi T, Akisawa A, Ng KC, Chua HT. Waste heat
driven dual-mode, multi-stage, multi-bed regenerative adsorption system.
International Journal of Refrigeration 2003;26:749e57.
[39] Qenawy AM, Mohamad AA. Current technologies and future perspectives in
solar powered adsorption systems. In: Canadian solar buildings conference,
Montreal; 2004.
[40] Corporation Methanex. Technical information & safe handling guide for
methanol. Version 3.0. See also:, http://www.methanex.com/environment/
technical.html; September 2006.
[41] Wu WD, Zhang H, Sun DW. Mathematical simulation and experimental study
of a modified zeolite 13X-water adsorption refrigeration module. Applied
Thermal Engineering 2009;29:645e51.
[42] Al-Mers A, Mimet A. Numerical study of heat and mass transfer in adsorption
porous medium heated by solar energy: Boubnov-Galerkin method. Heat
Mass Transfer 2005;41:717e23.
[43] U.S. Department of Commerce. National Oceanic & Atmospheric Administra-
tion. NOAA Research. Earth System Research Laboratory. Global Monitoring
Division. NOAA Solar Calculator.http://www.esrl.noaa.gov/gmd/grad/solcalc/
[44] Environment Canada. National Climate Data and Information Archive. www.
climate. Hourly Data Report for July 30, 2009. www.climate.weatheroffice.
gc.ca