ARTICLE IN PRESS

RENE3345_proof  20 March 2009  1/9

Renewable Energy xxx (2009) 19

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

2
3
4
5

55

Technical Note

Study of an adsorption refrigeration system powered by parabolic trough collector

and coupled with a heat pipe
a, *

, A. Mimet , M. Perez-Garca

A. El Fadar

8
9

Energetics Laboratory, Sciences Faculty, BP 2121, Abdelmalek Essaadi University, 93000 Tetouan, Morocco
b
Dpto. de Fsica Aplicada, Universidad de Almera, Spain

OO

6
7

10
11

18
19
20
21
22
23

Article history:
Received 8 April 2008
Accepted 8 March 2009
Available online xxx

The aim of the current paper is to propose a study of a novel solar adsorptive cooling system, using
activated carbonammonia pair, coupled with a parabolic trough collector (PTC) and a water-stainless
steel heat pipe. A theoretical model, based on the thermodynamics of the adsorption process, heat and
mass transfer within the porous medium and energy balance in the hybrid system components, is
developed and a simulation code, written in FORTRAN, is carried out. This model, which has been
validated by experimentation results, computes the temperature, pressure and adsorbed mass inside the
adsorbent bed. The performance is assessed in terms of specic cooling power (SCP) and solar coefcient
of performance (COPs). Furthermore, the effect of some important parameters on the system performance is discussed, and an optimization of these parameters is given.
The simulation results have shown that there exists, for each aperture width value of the collector (W),
an optimum external radius of adsorbent bed (R2). Under the operating and design conditions of
evaporation temperature Tev 0  C, condensing temperature Tcon 28  C, adsorption temperature
Tads 24  C, W 0.70 m, R2 0.145 m and reactor length of 0.5 m, an optimal corresponding COPs is
found to be of the order of 0.18.
2009 Published by Elsevier Ltd.

PR

16
17

a b s t r a c t

Keywords:
Solar energy
Parabolic trough collector
Heat pipe
Adsorption
Modelling
Refrigeration

ED

14
15

a r t i c l e i n f o

CT

12
13

24
25
26
27

34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54

During recent decades, an increasing attention has been paid to

the development of adsorption refrigeration technology due, on the
one hand, to the negative environmental consequences related to
conventional vapour compression refrigeration machines and, on
the other hand, to the benets of the adsorption systems whose
refrigerants are regarded as absolutely benign for the environment,
i.e. these refrigerants satisfy the Montreal protocol on ozone layer
depletion and the Kyoto protocol on global warming [1]. Moreover,
these systems have the benets of simpler control, silent operation
with no vibrations and lower operation costs, do not present
corrosion problems due to the working pairs normally used [2],
they could be operated with few moving parts, which reduce the
requirement for maintenance, and they do not need a solution
pump or rectier for the refrigerant like the absorption systems [3].
Additionally, solar adsorption refrigeration systems are attractive, mostly in remote areas without grid-connected electricity,
since solar radiation is freely available, and the cooling requirements increase particularly in the sunny regions.

CO

32
33

1. Introduction

UN

30
31

RR
E

28
29

* Corresponding author. Tel.: 212 66280523; fax: 212 39 99 45 00.

E-mail address: aelfadar@yahoo.fr (A. El Fadar).

56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84

The most commonly working pairs used in this eld are zeolite
water; active carbonmethanol; active carbonammonia; silica
gelwater. In this regard, some units of solar adsorption refrigerators have been commercialized using activated carbonmethanol
system manufactured by BLM Co. of France and zeolitewater
system manufactured by Zeopower Co. of USA. These units were
technically successful, but their costs are not competitive with the
conventional vapour compression system [4].
In addition to their high costs, adsorption refrigeration systems
have some other drawbacks, such as low specic cooling power and
low coefcient of performance, due to the weak heat transfer within
the adsorbers. Thus, in order to overcome these limitations and attain
higher performance, several approaches have been undertaken such
as enhancement of heat transfer properties in the adsorbent beds
[5,6], design and investigation of advanced cycles, and improvement
of regenerative heat and mass transfer between beds, [712].
On the other hand, the solar power technology employing the
parabolic trough collector (PTC) has been used in various applications, such as electricity generation, desalination, heating, etc.
Parabolic trough collectors are regarded as the most developed and
deployed type of solar concentrators [13] and their technology is
the most-veried solar technology through deployment and
construction testing [14]. Moreover, Bird and Drost [15] have

85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109

0960-1481/$ see front matter 2009 Published by Elsevier Ltd.

doi:10.1016/j.renene.2009.03.009

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

ARTICLE IN PRESS
2

RENE3345_proof  20 March 2009  2/9

A. El Fadar et al. / Renewable Energy xxx (2009) 19

175

110

113
114

123
124

A
Ac
C
Dgi
Dgo
Di
Do
Dwi
h
H
Ib
Lc

125
126

Le

127
128

L(Tev)

129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149

m
ma
P
Qc
q
r
R1
R2
t
tcycle
T
u
U
W
x

Greek letters
a
absorptivity

150
151

154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170

recommended that the PTC concept should receive the highest

priority for commercial development for low temperature (65
177  C) solar process heat applications. Even so, solar adsorption
cooling systems are usually carried out with at plate and evacuated
tube collectors, whereas much less works have been devoted to
concentrating collectors.
Besides, the use of heat pipe in solar collector is attractive
because it presents some advantages, such as construction
simplicity, perfect adjustability, easier control and high heat transfer ability in low temperature differences, ability to operate as
a thermal diode and as a thermal switch [16]. In addition, it does not
need moving parts or external pumping.
The objective of the current work is to present a study of an
adsorption refrigeration system powered by solar energy, in which
the parabolic trough collector is combined with a water-stainless
steel heat pipe, in an attempt to improve the performance of
intermittent refrigeration systems. A numerical analysis is performed to examine the optimal parameters of the system.

UN

152
153

171
172

2. System description

173
174

The simplied scheme of the hybrid system is shown in Fig. 1, it

comprises a parabolic trough collector (PTC), a heat pipe (HP),

176
177
178
179
180
181
182
183
184
185
186
187
188
189

Subscripts
a
adsorbed, adsorbate
ab
absorber
ads
adsorption
amb
ambient
con
condensation, condenser
e
equivalent
ev
evaporation, evaporator
g
gas, glass
gl
global
hp
heat pipe
l
liquid
s
solid (activated carbon)
sat
saturation
w
wick
wa
wall (stainless steel)

190
191

Abbreviations
AC
activated carbon
COPs
solar coefcient of performance
HP
heat pipe
PTC
parabolic trough collector
SCP
specic cooling power (W kg1)

207
208

DP

121
122

3
3w
r
s
q
DHads
Dx

collector optical factor

reectivity of the reecting surface
external metallic envelope thickness of the reactor
(m)
bed porosity
porosity of wick
density (kg m3)
transmittivity
volume fraction of adsorbed phase
isosteric heat of adsorption (J kg1)
adsorption capacity difference between adsorption
and desorption phases (kg kg1)
thermal conductivity (W m1 K1)

TE

119
120

EC

117
118

CO
RR

115
116

cross-sectional area (m2)

collector aperture area (m2)
specic heat (J kg1 K1)
inner diameter of glass tube (m)
outer diameter of glass tube (m)
container inner diameter of absorber (m)
container outer diameter of absorber (m)
wick inner diameter of heat pipe (m)
heat transfer coefcient (W m2 K1)
specic enthalpy of ammonia (J kg1)
direct solar radiation (W m2)
condenser section length of heat pipe/reactor length
(m)
evaporator section length of heat pipe/collector length
(m)
latent heat of ammonia at evaporation temperature
(J kg1)
mass (kg)
adsorbed mass of ammonia on a layer of activated
carbon (kg)
pressure (bar)
cooling production (J)
ow rate (kg s1)
radial coordinate (m)
internal radius of adsorbent bed (m)
external radius of adsorbent bed (m)
time (h)
cycle time (h)
temperature (K)
specic internal energy (J kg1)
overall heat loss coefcient (W m2 K1)
aperture width of collector (m)
adsorbed mass of ammonia per unit mass of adsorbent
(kg kg1)

b
g
d

OF

Nomenclature

RO

111
112

192
193
194
195
196
197
198
199
200
201
202
203
204
205
206

209
210
211
212
213
214
215
216

a cylindrical adsorber containing activated carbon, an air condenser,

an evaporator, an ammonia liquid tank, and refrigerant valves.
The evaporator end of the heat pipe is placed at the focused line
of the solar collector, while its condenser is inserted in the adsorbent bed. The reective surface of the PTC focuses the solar direct
radiation on the linear absorber (receiver). Then, the solar radiation
is converted to thermal energy, which is absorbed at the heat pipe
evaporator section. This heat vaporizes the working uid (water) in
this section. The resulting difference in pressure drives vapour from
the evaporator to the condenser of HP, where it is cooled down and
condensed, and then the latent heat of vaporization is released to
the heat sink (adsorbent bed). The capillary pressure, which is
generated by the wick, drives the condensed liquid to the evaporator for re-evaporation. The adsorption and desorption processes
are generated by alternate cooling and heating of the adsorber, as
follows:
During daytime, when the heating phase begins, solar energy
heats the adsorber to the maximum cycle temperature, and when
the pressure inside the adsorber reaches the pressure in the
condenser (saturation pressure at the temperature of the
condenser), the ammonia vapour starts desorbing from the adsorbent bed and it is then condensed in the condenser; heat is dissipated to the surroundings and the condensate is collected in a tank.

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239

ARTICLE IN PRESS

RENE3345_proof  20 March 2009  3/9

A. El Fadar et al. / Renewable Energy xxx (2009) 19

305

241
242

306
307

243
244

308
309

245
246

310
311

247
248

312
313

249
250

314
315

251
252

316
317

OF

240

253
254
255
256

261
262

Fig. 1. (a) Simplied scheme of the hybrid system: 1. PTC; 2. Heat pipe; 3. Adsorber; 4. Evaporator; 5. Condenser; 6. Valves; 7. Ammonia liquid tank; 8. Expansion valve. (b) Cross
section of the receiver assembly: 9. Reective surface; 10. Glazing; 11. Absorber; 12. Wick.

DP

259
260

RO

257
258

263

270
271

3. Mathematical modelling

272
273

3.1. Model assumptions

278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304

EC

276
277

The main model assumptions adopted in this work are as

follows:
i. The pressure is uniform inside the adsorbent bed.
ii. The adsorbent bed is considered as a continuous medium. The
conduction heat transfer in the medium can be characterised
by an equivalent thermal conductivity, le.
iii. The adsorption/desorption process is an isobaric process.
iv. The porous medium properties have a cylindrical symmetry.
v. All phases are continuously in thermal, mechanical and
chemical local equilibrium.
vi. The heat transfer in adsorbent is radial and the convection
heat transfer due to the radial mass transfer is neglected.
vii. A steady-state heat pipe operation is supposed.
viii. Heat pipe temperature is assumed to be equal to liquid
vapour interface temperature.
ix. The vapour temperature is uniform along the heat pipe.
x. The glass tube temperature is uniform at any given time.

CO
RR

274
275

Some assumptions above have to be claried: rst, regarding the

uniform-pressure assumption, several researches presented in
literature assumed uniform-pressure distribution inside the adsorbent bed [1719]. This point has also been investigated by Sun et al.
[20], they studied the coupled heat and mass transfers for two
adsorbent/adsorbate pairs: zeolite 13X-water and 13X-ammonia,
whose operating pressures are considerably different, they
concluded that the zeoliteammonia pair, which operates at much
higher pressures (several bar) and the corresponding pressure
variations in the bed are, in general, negligible, the uniform-pressure assumption is acceptable. Also, the experiments carried out

320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354

UN

266
267

with activated carbon/ammonia [21], have shown that the pressure

drop in the bed (DP) is of the order of 14.6  103 Pa, and the relative
error (DP/P) is of the order of 108 when neglecting the pressure
drop.
Concerning the isobaric desorption/adsorption assumption, in
real cycles, pressure is not a constant during desorption and
adsorption processes because of the non-equilibrium of these
processes. With this assumption, the inuences on the results are, in
general, more pronounced for desorption phase than that of
adsorption, this is because in the former case, the bed temperature
is more inuenced by the atmospheric conditions (solar radiation,
ambient temperature, wind velocity, etc.) and also, the condenser
temperature changes, due to the ambient temperature uctuation
in the daytime. Thus, because of the small variations of temperature
in the adsorber during the nighttime, the adsorption process
remains close to that of an isobar. Nevertheless, It is noted that in
the current work, the simulation results are obtained with climatic
data of a summer day (see Fig. 2), where the ambient temperature
uctuates within a narrow range (between 24 and 28  C). So, this
would not cause a high variation of the condensation temperature.

TE

268
269

During the cooling period, the adsorbent bed is cooled down to

ambient temperature, therefore its pressure decreases, and when
this pressure is equal to the evaporation pressure (the saturated
vapour pressure of ammonia), the refrigerant boils in the evaporator. The resulting ammonia vapour is re-adsorbed into the
adsorbent, whereas cooling effect is produced.

264
265

318
319

355
356
357
358
359
360
361
362
363
364
365
366
367
Fig. 2. Comparison of predicted and measured temperatures.

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

368
369

383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434

leff

l lw  1  3w ll  lw 
ll l
ll lw 1  3w ll  lw 

(1)

where ll and lw denote the thermal conductivities of the liquid

phase of the working uid (water) and wick material, respectively,
while 3w designates the porosity of the wick.
3.2. Model equations

According to the assumptions stated above, a one-dimensional

heat conduction model is used for the heat pipe (wall and liquidwick regions) and a one-dimensional transient model based on heat
and mass transfer and the thermodynamics of the adsorption
process inside the reactor, are suggested.
The different equations of this model are expressed as below:
3.2.1. Glazing envelope
The energy balance equation of the glass tube is written as:

rg Cg Ag



vTg
gag bWIb t pDgi Uabg Tab  Tg
vt


 pDgo Ugamb Tg  Tamb

(2)

3.2.2. Energy balance in the absorber

The energy balance equation in the absorber (receiver) is
expressed as follows:



vT
rab Cab Aab ab gsaab bWIb t  pDo Uabg Tab  Tg
vt


 pDo hT Tab  Thp

RT 1=2pLe

lwa

(6)

where Lc designates the adsorber length.

(3)

#
lnDi =Dwi

leff

In Eqs. (2) and (3), Ugamb and Uabg denote the heat transfer
coefcients between the glazing envelope and ambient air, and
between the absorber and glazing envelope, respectively. The
convective and radiative heat transfer coefcients are obtained from
correlations given in the literature [2628].
3.2.3. Heat and mass transfer equations in the adsorbent bed
3.2.3.1. Mass conservation equation. Heat transfer within the
porous medium induces a mass transfer and vice versa. Indeed, the

438
439
440
441
442
443
444
445

448
449
450
451
452
453
454
455
456
457
458
459
460

X
ddU X

qout Hout 
qin Hin f z
dt
out

463
464

(7)

in

where Hin and Hout are the enthalpies of adsorbate entering and
leaving the layer under consideration, respectively; qin and qout
designate the mass ow rates of adsorbate entering and leaving the
same layer, respectively; f represents heat ux exchanged with
surrounding medium, it is expressed according to the law of Fourier,
while z refers to mechanical power exchanged with surrounding
medium which is nil in the present case.
The differential internal energy, dU, in Eq. (7) may be expressed
as a function of different phases (i.e. solid, gaseous and adsorbed
phases) in the porous reactive medium as:

i
h
dU 1  3rs us 3  qrg ug qra ua

(8)

i
vh 
Vl 1  3rs us 3  qrg ug qra ua
vt
qr dr; tHg Tr dr; p
#
"
v2 T 1 vT
 qr; tHg Tr; p le Vl

vr 2 r vr

461
462

465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484

485
486
487
488

where

(4)

436
437

3.2.3.2. Energy balance equation. The energy balance equation is

obtained by applying the rst law of thermodynamics to an open
system (the layer under consideration):

Thus, Eq. (7) can be formulated as follows:

where hT is the heat transfer coefcient between the outer surface

of absorber and the liquidvapour interface. The corresponding
thermal resistance, RT, is given by:

"
lnDo =Di

(5)

where qr; t is the ammonia ow rate within the layer under

consideration; 3 is the porosity of adsorbent; q is the volume fraction
of adsorbed phase, and Vl denotes the volume of this layer which is
given by:

Vl 2prdrLc

435

446
447

i
vh 
vqr; t
V 3  qrg qra
dr
qr; t  qr dr; t 
vt l
vr

OF

381
382

RO

379
380

DP

377
378

hotter layers in adsorbent bed desorb the ammonia gas, which will
be adsorbed on colder layers. This mass transfer contributes to the
heat transfer process, since the gas that will be adsorbed on cold
layers is hot, and also because of the exothermal character of
adsorption.
Therefore, if we consider a layer of radial coordinate, r, and
thickness, dr, in the adsorbent bed, the difference between mass
ow rates of ammonia gas entering and leaving this layer is equal to
the adsorbed and gaseous ammonia remaining in the same layer,
per unit of time. The ammonia mass conservation equation is
expressed as follows:

TE

375
376

Furthermore, the isobaric assumption is widely adopted in literature and, under some working conditions, this assumption was
found to be consistent with the experimental and simulated results
[22,23].
Regarding the uniformity of vapour temperature along the HP,
the validation of this assumption has been performed experimentally by Huang and El-Genk [24]; they measured the axial vapour
temperature distributions of a copperwater heat pipe. As for the
assumption relating to the convection heat transfer, we will discuss
this point later (Section 3.2.3). The other assumptions have been
made to simplify the model; they are commonly adopted in literature. Consequently, we can conclude that the assumptions made in
this work could be considered as acceptable.
]Besides, under the condition of the metal screen wick, which
is assumed to be saturated with the working uid, the effective
wick thermal conductivity is evaluated by the following formula
[25].

EC

373
374

RENE3345_proof  20 March 2009  4/9

A. El Fadar et al. / Renewable Energy xxx (2009) 19

CO
RR

371
372

UN

370

ARTICLE IN PRESS

(10)

489
490

In Eq. (9), Tr represents the temperature of bed at radial

position r; p denotes the bed pressure; DHads is the adsorption heat,
whereas Hg P; T and Ha P; T designate the specic enthalpies of
ammonia at gaseous and adsorbed phases, respectively.

491

3.2.3.3. Combined heat and mass transfer equation in the adsorbent

bed. The heat and mass transfer equation in the adsorbent bed is
obtained by combining the energy balance and mass conservation
equations, i.e. Eqs. (5) and (9):

496
497

Hg P; T Ha P; T DHads

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

492
493
494
495

498
499

503
504
505
506
507
508
509
510
511
512
513
514
515
516

A. El Fadar et al. / Renewable Energy xxx (2009) 19


vug
vus
v
us 1  3rs 3  qrg
2prLc 1  3rs
vt
vt
vt



v
v
vua
ug  Hg 3  qrg ua  Hg qra qra
vt
vt
vt
#
"
!
vHg
v2 T
1 vT
11
qr; t
2pr le Lc

r vr
vr
vr 2
r

If we adopt the approach that considers molecules in the

adsorbed phase (liquid) as incompressible, the differential internal
energy for both solid (activated carbon) and adsorbed phases can be
written as:

du Cv dT Cp dT

(12)

The specic enthalpy may be expressed as function of internal

energy, pressure and density as:

517
518

ug  Hg 

519
520

and, by using Eq. (10), we obtain:

521
522

ua  Hg 

 DHads

(14)

then, Eq. (11) becomes:

526
527

532
533
534
535

540
541
542
543
544
545
546
547
548
549
550
551

where ma is the adsorbed mass of ammonia on the layer considered

in activated carbon, it is related to the volume fraction of the
adsorbed phase, q, by:

CO
RR

538
539

ma qra Vl

552
553

vma
q=Vl DHads
vt

554
555

DHads 1=Vl

556

q
vT
q vT
q vHg
cg dr
dr q=Vl dHg
cg
2prLc vr
Vl vr
Vl vr

557
558
559
560
561
562
563
564

(16)

The different terms of Eq. (15) represent, respectively, (I) the

energy for heating activated carbon and ammonia at gaseous and
adsorbed phases; (II) elastic energy of ammonia gas; (III) elastic
energy of adsorbed ammonia; (IV) heat of adsorption; (V)
heat exchanged by convection; and (VI) heat exchanged by
conduction.
However, the convective term (V) can be neglected when
compared with term (IV). Indeed, by expanding these two terms, we
can put in evidence their comparison:

UN

536
537

(17)

Thus, based on the fact that the temperature difference between

two consecutive layers in the bed is at most of the order of 2  C,
which corresponds to a weak variation of specic enthalpy, dHg ,
(e.g.dHg varies roughly 1 kJ/kg of NH3 at 30  C), whereas the variation in adsorption heat, DHads , is of the order of 1500 kJ/kg of
NH3 [21].

566
567
568
569

19

3.2.4. Boundary and initial conditions

The initial and boundary conditions are listed below:

570
571

574
575
576
577

- Boundary conditions:

r R1




vT
 le
hgl Thp  T
vr rR1

r R2

 le

(20)

R2 < r < R2 d

rC=lwa

vTwa
vt

v2 Twa 1 vTwa

r vr
vr 2

578
579
580
581


vT
hi T  Twa r R2 ; t
vr rR2

(21)

582
583
584
585

!
(22)

586
587
588
589
590
591
592
593
594
595
596

(15)

597
598
599
600

r R2 d

vTwa
vr

601
602

(23)

rR2 d

where hgl is the global heat transfer coefcient between the heat
pipe vapour and the adsorbent bed. The equation of corresponding thermal resistance, Rgl , is shown in Eq. (24):

Rgl 1=2pLc

"
lnDo =Di

lwa

lnDi =Dwi

leff

hi R1

#
(24)

where hi is the internal heat transfer coefcient between the

adsorbent bed and metallic wall; le is the equivalent thermal
conductivity of the bed; hi andle were evaluated experimentally
[21].
- Initial conditions:

Tt 0 Tab t 0 Tg t 0 Tads ;
(18)

565

572
573

TE

530
531

#
"
h
ivT
v2 T 1 vT
3
r
3
q
r
qr
l
e
1  s Cs  g Cg a Ca

vt
vr 2 r vr



p
v
1 p
vma

3  qrg

DHads
rg Vl ra
vt
vt

z}|{
z}|{
z}|{
z}|{
z}|{
h
ivT
v


vma III
vma IV
q
vT
I
II
3  qrg
1  3rs cs qra ca 3  qrg cg
 p=rg
 1=Vl p=ra
 DHads 1=Vl

cg V
vt
vt
2prLc vr
vt
vt
z}|{
#
"
v2 T 1 vT VI
le

vr 2 r vr

EC

528
529

Consequently the heat and mass transfer equation in the

adsorbent bed is expressed as follows:

DP

ra

523
524
525

(13)

rg

OF

501
502

RENE3345_proof  20 March 2009  5/9

RO

500

ARTICLE IN PRESS

Pt 0 Pev

Psat Tev

Furthermore, to complete the mathematical formulation of the

model, the following data are needed:
1) Adsorption equilibrium model: The adsorption equilibrium data
of the considered adsorbent/adsorbate pair (activated carbon/

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629

A. El Fadar et al. / Renewable Energy xxx (2009) 19

a correlation available in the literature [33]. The climatic data are

illustrated in Fig. 2. The other input data to the simulation
program are given in Table 1.

Table 1
Main input data used in the simulation.
Symbol Parameter

Value

Unit

633
634

Parabolic trough collector

Cab
Specic heat capacity of the absorber

0.49

kJ kg1 K

635
636

Cg

Specic heat capacity of glass tube

0.75

kJ kg1 K

Dgi
Dgo
Do
Le

Inner diameter of the glass tube

Outer diameter of the glass tube
Outer diameter of the absorber
Collector length
Absorptivity of the absorber
Absorptivity of the glass tube
Collector optical factor
Emissivity of the absorber
Emissivity of the glass tube
Density of the absorber
Density of the glass tube
Reectivity of the reecting surface
Transmittivity of the glass tube

0.11
0.115
0.056
1.00
0.92
0.05
0.90
0.90
0.85
7850
2500
0.90
0.90

m
m
m
m

kg m1
kg m1

1.00
0.048
0.044
0.73
0.63

m
m
m

W m1 K

639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655

1

aab
ag
b
3ab
3g
rab
rg
g
s

Heat pipe (HP)

Le
Evaporator length
Di
Container inner diameter of HP
Dwi
Wick inner diameter of HP
3w
Porosity of wick
ll
Thermal conductivity of the liquid phase of the working
uid
lw
Thermal conductivity of wick material

COPs

1

Wm

656
657

Activated carbon/ammonia pair

Cs
Specic heat of adsorbent

0.836

kJ kg1

658
659

hi

33.45

W m2 K

664
665
666
667
668
669

L (Tev)
R1

3
le

Operating conditions
Tads
Adsorption temperature
Tcon
Condensation temperature
Tev
Evaporation temperature

676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694

7
Cl dT 5

1600
kJ kg1
1262.40 kJ kg1

Tev

 The SCP is dened as the ratio between the cooling production

and the cycle time per unit of adsorbent mass:

0.028
0.71
0.431

W m1 K

SCP

297
301
273

K
K
K


x x0

cpr
1  pr

1  n 1pnr npn1
r
1 c  1pr  cpn1
r

(25)

where x0 is the adsorbate mass, which is necessary to form

a monomolecular layer; pr p=psat is the relative pressure,
where p and psat are equilibrium pressure and saturation pressure of the adsorbate, respectively; c is a constant, that characterizes the adsorbate and n denotes the number of molecular
layers of the adsorbate. For the adsorbent/adsorbate pair used in
this work, the numerical values of c, n and x0 have been taken
from literature [21]; c 3.20; n 1.62; x0 0.271 kg kg1.
2) Thermodynamics equilibrium properties of the reactive medium
components are taken from the literature, namely, activated
carbon [29], ammonia gas [30] and adsorbed ammonia [31].
3) Global solar irradiance and ambient temperature that we have
used in simulation correspond to the case of a clear type day of
July, measured in Tetouan (35 350 N, 5 230 W), north of Morocco,
with Mediterranean climate [32]. The hourly diffuse radiation
has been calculated from the hourly global radiation, using

702
703
704
705

708
709

Qc
tcycle mAC

712
713
714
715
716
717
718

(27)

1

ammonia) are described by the Brunauer Emmett Teller (BET)

theory, which allows one to evaluate the adsorbed mass of
ammonia per unit mass of adsorbent, x, as follows:

UN

674
675

ZTcon

(28)

719
720
721
722
723
724
725
726
727
728

4. Numerical solution

729
730

In this study, an implicit nite difference scheme is used for

solving the above mathematical model numerically. The discretized
equations are solved using the Tri-Diagonal Matrix Algorithm
(TDMA) and the non-linearity of the equations is solved by iterative
techniques.
A computer program, written in FORTRAN, based on this
numerical scheme has been developed in order to simulate the
behaviour of the cooling system. It computes, for each value of time,
the temperature, pressure and adsorbed mass inside the reactor.
The performance parameters have been evaluated as well.

731
732

1

670
671
672
673

700
701

710
711

Ac Ib tdt

6
Qc mAC Dx4LTev 

TE

DHads

(26)

where Qc is the cooling effect produced in the evaporator, which is

equal to the refrigerant latent heat of evaporation minus the
sensible heat to cool down the refrigerant from the condensation
temperature to the evaporation temperature, it is expressed as:

EC

662
663

1

CO
RR

660
661

Qc

sunset
Z

696
697

706
707

sunrise

1

Heat transfer coefcient between the HP wall and the

adsorbent bed
Latent heat of adsorption
Latent heat of ammonia at evaporation temperature
(0  C)
Internal radius of adsorbent bed (m)
Porosity of adsorbent bed
Equivalent Thermal conductivity of adsorbent bed

The assessment parameters of the performance of the adsorption refrigeration system considered in this work are the solar
coefcient of performance (COPs) and the specic cooling power
(SCP). These parameters are dened by the following formulas:
 The solar coefcient of performance is expressed as:

1

46.00

3.3. System performance

DP

637
638

1

695

698
699

OF

631
632

RENE3345_proof  20 March 2009  6/9

RO

630

ARTICLE IN PRESS

5. Experimental validation
In order to validate the model of heat and mass transfer within
the porous medium, described above, a tubular reactor with a double
stainless steel envelope heat exchanger has been used. This reactor
has an inner diameter and a length equal to 53 mm and 250 mm,
respectively; it was heated at isosteric phase by means of thermal oil,
which circulates along the 4 mm thickness of the inter-envelope
space with a ow rate of 1.5 l/min. The thermostat temperature
ranged between 20 and 250  C, and each reactor cover contains, in its
centre, a hole allowing the inlet or outlet of the ammonia gas. The
reactor was packed with 274 g of activated carbon (AC) particles
having a mean diameter of 2 mm. The type of AC used in the experience is BPL 4  10 from CHEMVIRON rm [29]. Detailed information about this experimental device is shown in a previous work [21].
In Fig. 3, we have reported experimental and model results; this
gure depicts the time evolution of predicted and experimental
temperatures at three different radial positions inside the reactor

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759

ARTICLE IN PRESS

RENE3345_proof  20 March 2009  7/9

A. El Fadar et al. / Renewable Energy xxx (2009) 19

825

761
762

826
827

763
764

828
829

765
766

830
831

767
768

832
833

769
770

834
835

771
772

836
837

OF

760

773
774
775
776

RO

777
778
779
780
Fig. 3. Climatic data used in simulation.

783

786
787
788
789

(2,3,6). The temperatures were measured using six thermocouples

placed in different points of the cylindrical adsorbent bed. See
Table 2.
The comparison between these values shows an excellent
agreement. Therefore, the one-dimensional model of heat and mass
transfer within the adsorbent bed is validated.

790
791

798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820

EC

796
797

In this section, we illustrate the simulation results obtained with

computation program. It is noted that, because of the characterizations of the activated carbon/ammonia pair, we target system
operation at moderate temperatures, in the range of 70170  C.
Fig. 4 depicts the time evolution of the adsorbent bed temperature at three different radial positions (layers) within the adsorber.
It shows that there is a temperature gradient between these positions, and that the temperature prole of each layer is strongly
inuenced by the solar radiation uctuation.
Fig. 5 represents the variation of both adsorbed mass of
ammonia per unit of adsorbent mass and pressure inside the
adsorber during the cooling system cycle, which comprises of four
phases, namely, isosteric heating, isobaric condensationdesorption, isosteric-cooling and isobaric evaporationadsorption. These
phases correspond to the diurnal and nocturnal solar radiation
periods. It is noteworthy to mention that, even though the adsorbed
amount remains constant during the isosteric phases (heating and
cooling), its distribution inside the adsorbent bed is not uniform,
due to the desorption and adsorption processes that could partially
occur between the hottest layers and the coldest ones.
In Fig. 6, we have illustrated the effect of the adsorbent mass on
the performance of adsorption cooling system. Thus, it can be seen
that the specic cooling power (SCP) and solar coefcient of
performance (COPs) are much sensitive to adsorbent mass variation.
The SCP decreases with the increase in adsorbent mass. This is

CO
RR

794
795

6. Results and discussion

840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874

UN

792
793

because, with a higher value of adsorbent mass, the thermal resistance increases. Consequently, the heat exchange rate during the
cycle process is reduced contributing to a longer cycle time. As
result, a reduction in the SCP. So, it is obvious that the compactness Q1
attainment of the system requires a low mass of activated carbon.
The COPs increases with the increase in adsorbent mass, and
once the adsorbent mass reaches a critical value, which is evaluated
to be 14.5 kg, the COPs decreases. The reason is that the increase in
adsorbent mass induces adsorption of high quantity of ammonia
vapour mass initially at adsorption phase and hence, desorption of
large amount of ammonia at subsequent desorption process. This
produces more cooling and, consequently, results in high COPs.
Nevertheless, beyond this optimal value of 14.5 kg, the adsorbent
bed is heated, but the heat absorbed is not sufcient to desorb the
required amount of ammonia. For a given aperture width of

TE

784
785

Fig. 4. Temperature prole in different layers within the adsorbent bed (R2 0.12 m;
W 0.84 m). Layer 1: adjacent layer to heat pipe (r 2.9 cm); Layer 2: layer in the
middle (r 7.4 cm); Layer 3: adjacent layer to external envelope (r 11.9 cm).

DP

781
782

838
839

875
876
877
878
879
880
881
882
883
884
885

Table 2
Thermocouples positions within the cylindrical adsorber.

821
822

Thermocouple number, i

823
824

Radial position, ri (cm)

Axial position, zi (cm)

0.00
6.00

0.00
12.00

1.5
12.00

2.65
12.00

0.00
18.00

1.00
18.00

886
887
Fig. 5. Variation of pressure and adsorbed mass within the reactor, during a simulated
cycle (R2 0.12 m; W 0.84 m).

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

888
889

ARTICLE IN PRESS
8

RENE3345_proof  20 March 2009  8/9

A. El Fadar et al. / Renewable Energy xxx (2009) 19

955

890
891
892
893
894

COPs

895
896
897
898
899
900

0.15

958
959

0.1

960
961

0.05

962
963

0
1

964
965

905
906

922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945

External radius of adsorbent (m)

947
948
949
950
951
952
Fig. 7. Effect of the aperture width of collector on COPs.

RO

966
967
968
969
970
971
972
973
974
975

In this paper, a theoretical model has been established to

investigate the performance of a novel adsorption refrigerating
system driven by parabolic trough solar collector. This system is
coupled with an annular heat pipe used for transferring heat
towards the adsorbent bed.
An experimental device has been utilised in order to check the
correctness of the heat and mass transfer model within the adsorbent bed. The comparison revealed a good agreement between
model predictions and experimental data. A numerical program has
been developed to examine the effect of some design parameters on
the performance of adsorption cooling system at given operating
conditions. From the analysis performed above, some conclusions
are summarized as follows:

976
977

DP

946

953
954

0.1

0.25

TE

920
921

EC

918
919

collector of 72.8 cm, a maximum COPs of 0.18 is achieved for an

adsorbent mass of 14.5 kg.
It is also observed from Fig. 7 that, at given value of external
radius of adsorbent bed, R2, (e.g. R2 14 cm, in this gure) and in the
range investigated [33.684 cm] of aperture width of collector (W),
the COPs increases with W. This is because, when W increases, the
solar radiation collected becomes higher, and more heat is absorbed
by the reactor, hence higher amount of ammonia is desorbed, as
result, more cooling effect is produced. However, when W reaches
roughly 0.70 m, the COPs varies slightly, this can be interpreted by
the fact that, increase in W causes only heating (increase in sensible
heat) of the adsorber (adsorbent bed and metallic part) and, beyond
this optimal value, a small quantity of ammonia is desorbed. So,
values of W, higher than 70 cm, are not recommended for this
system, when R2 14 cm.
Fig. 8 depicts the variation of COPs with both W and R2. From this
gure, it is observed that, for each aperture width value, there exists
an optimum external radius of adsorbent bed, which corresponds to
a maximum COPs.

CO
RR

916
917

0.2 0.05

0.2

7. Conclusions
Fig. 6. Inuence of the adsorbent mass on SCP and on COPs, (W 0.728 m).

UN

914
915

0.4

0.15

Fig. 8. COPs variation with aperture width of collector and external radius of adsorbent
bed.

907
908

913

0.6

Aperture width (m)

903
904

911
912

0.8

OF

901
902

909
910

956
957

0.2

1. The simulation results showed that both adsorbent mass and

collector area (aperture width) have major inuences on the
system performance. These results indicated also that, for each
value of aperture width of collector, there is an optimum choice
for reactor dimensioning.
2. In the ranges of investigation, it was put in evidence that the
system is optimum (COPs 0.18), when the adsorber external
radius and aperture width of collector are of the order of 14.5
and 70 cm, respectively. So, the results show a promising
performance in comparison with published data, which were
obtained with intermittent adsorption refrigerators.
3. In addition to its high efciency, due to high efciency of the
parabolic trough collector and high ux density of the heat pipe,
the proposed system presents also the advantage of lightness,
when compared with those that use at plate or evacuated tube
collectors.
References
[1] Anyanwu EE. Environmental pollution: restructuring the refrigeration industry
as a way out. Int J Environ Protect Eng 2000;26(4):1727.
[2] Wang RZ, Oliveira RG. Adsorption refrigeration an efcient way to make good
use of waste heat and solar energy. Prog Energy Comb Sci 2006;32:42458.
[3] Oliveira RG, Silveira Jr V, Wang RZ. Experimental study of mass recovery
adsorption cycles for ice making at low generation temperature. Appl Thermal
Eng 2006;26:30311.
[4] Alghoul MA, Sulaiman MY, Azmi BZ, Wahab MAbd. Advances on multi-purpose
solar adsorption systems for domestic refrigeration and water heating. Appl
Therm Eng 2007;27:81322.
[5] Poyelle F, Guilleminot JJ, Meunier F. Experimental tests and predictive model of
an adsorptive air conditioning unit. Ind Eng Chem Res 1999;38:298309.
[6] Tamainot-Telto Z, Critoph RE. Adsorption refrigerator using monolithic carbon
ammonia pair. Int J Refrig 1997;20(2):14655.
Q2

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019

RENE3345_proof  20 March 2009  9/9

A. El Fadar et al. / Renewable Energy xxx (2009) 19

RO

OF

[20] Sun LM, Ben Amar N, Meunier F. Numerical study on coupled heat and mass
1045
transfers in an adsorber with external uid heating. Heat Recov Syst CHP
1046
1995;15:1929.
1047
[21] Mimet A. Etude theorique experimentale dune machine frigorique a`
adsorption dammoniac sur charbon actif, Ph.D. thesis, FPMs, Mons, Belgium;
1048
1991.
1049
[22] Critoph RE. An ammonia carbon solar refrigerator for vaccine cooling. Renew
1050
Energy 1994;5(1):5028.
1051
[23] Tamainot-Telto Z, Critoph RE. Adsorption refrigerator using monolithic carbon
ammonia pair. lnt J Refrig 1997;20:14655.
1052
[24] Huang L, El-Genk MS. Experimental investigation of transient operation of
1053
a water heat pipe. In: Proceedings of 10th symposium on space nuclear power
1054
and propulsion, Albuquerque, NM; 1993. p. 36574.
[25] Faghri A. Heat pipe science and technology. USA: Taylor & Francis; 1995.
1055
[26] Bejan A. Convection heat transfer. 2nd ed. New York: John Wiley & Sons; 1995.
1056
p. 32324.
1057
[27] Nafey S, Fath HS, El-Helaby SO, Soliman AM. Solar desalination using humidication dehumidication processes. Part I. A numerical investigation. Energy
1058
Convers Manage 2004;45:124361.
1059
[28] Mullick SC, Nanda SK. An improved technique computing the heat loss factor
1060
of a tubular absorber. J Solar Energy 1989;42:17.
[29] CHEMVIRON. Granular activated carbon. Bruxelles: Chemviron; 1988.
1061
[30] Institut International de froid. Tables et diagrammes pour lindustrie du froid,
1062
proprietes thermodynamiques du R12, R22, R717. Paris: Institut International
de Froid; 1981.
Q41063
[31] Mahamane A. Etude de ladsorption de vapeurs purs sur solides poreux, Ph.D.
1064
thesis, FPMs, Mons, Belgium; 1989.
1065
[32] Aroudam H. Evaluation du gisement solaire dans la region de Tetouan, the`se
1066
de 3e`me cycle, Faculte des sciences de Tetouan, Morocco; 1992.
1067
[33] Aroudam H, El Hammouti M, Ezbakhe H. Determination of correlations of solar
radiation measured in Tetouan. Renew Energy 1992;2:4736.
1068

CO
RR

EC

TE

DP

[7] Pons M, Poyelle F. Adsorptive machines with advanced cycles for heat pumping
or cooling applications. Int J Refrig 1999;22:2737.
[8] Critoph RE. Forced convection adsorption cycle. Appl Therm Eng
1998;18(3):28998.
[9] Meunier F. Theoretical performance of solid adsorbent cascading cycles using
zeolitewater and active carbonmethanol pairs: four case studies. Heat
Recovery CHP Syst 1986;6(6):4918.
[10] Wang RZ. Performance improvement of adsorption cooling by heat and mass
recovery operation. Int J Refrig 2001;24:60211.
[11] Douss N, Meunier F. Experimental study of cascading adsorption cycles. Chem
Eng Sci 1989;44:22535.
[12] Saha BB, Koyama S, Kashiwagi T, Akisawa A, Ng KC, Chua HT. Waste heat driven
dual-mode, multi-stage, multi-bed regenerative adsorption system. Int J Refrig
2003;26:74957.
[13] Badrana Omar, Eckb Markus. The application of parabolic trough technology
under Jordanian climate. Renew Energy 2006;31:791802.
[14] Price Hank, Hassani Vahab. Modular trough power plant. Cycle Syst Anal
January 2002. NREL/TP-55031240.
[15] Bird SP, Drost MK. Assessment of generic solar thermal concept for large
industrial process heat applications. In: Proceedings of the ASME solar energy
division, fourth annual conference, Albuquerque, NM; 1982.
[16] Abu-Zour A, Riffat S, Gillott M. New design of solar collector integrated into
solar louvres for efcient heat transfer. Appl Therm Eng 26(16):187682.
[17] Guilleminot JJ, Meunier F, Pakleza J. Heat and mass transfer in a non-isothermal
xed bed solid adsorbert reactor: a uniform pressure non-uniform temperature case. Int J Heat Mass Transfer 1987;30:1595606.
[18] Hajji A, Khallou S. Theoretical and experimental investigation of a constantpressure adsorption process. Int J Heat Mass Transfer 1995;38:334958.
[19] Sun M, Feng Y, Pons M. Numerical investigation of adsorptive heat pump
systems with thermal wave heat regeneration under uniform-pressure
conditions. Int J Heat Mass Transfer 1997;40:28193.

UN

1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037 Q3
1038
1039
1040
1041
1042
1043
1044

ARTICLE IN PRESS

Please cite this article in press as: Fadar AE, et al., Study of an adsorption refrigeration system powered by parabolic trough collector and coupled
with a heat pipe, Renewable Energy (2009), doi:10.1016/j.renene.2009.03.009

1069