Renewable Energy 28 (2003) 249–269

www.elsevier.com/locate/renene

Simulation and economic analysis of a solar-

powered adsorption refrigerator using an
evacuated tube for thermal insulation
C.H. Li, R.Z. Wang ∗, Y.J. Dai
Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, 1954 Hua Shan Road,
Shanghai 200030, China

Received 8 July 2001; accepted 8 February 2002

Abstract

A detailed analysis and simulation of heat transfer of an evacuated tube for a solar-powered
adsorption refrigerator, including inhomogeneous radiation heat transfer between its inner and
outer tubes, two adjacent tubes in the group, the tube and back plate, sunlight reflection on
back plate, etc. has been carried out. Also, a new cost–effectiveness parameter is provided to
assess the comprehensive performance of the system, and consequently optimums for the diam-
eter of tubes and the distance between two adjacent tubes are investigated.  2002 Elsevier
Science Ltd. All rights reserved.

Keywords: Evacuated tube; Solar energy; Adsorption; Economic analysis; Heat transfer

1. Introduction

In the last 20 years, gratifying achievements in technology for solid adsorption

refrigeration have been made. A variety of cycles [1], such as continuous cycle,
cycle with heat and mass recovery, cascade cycle and regeneration cycle, have been
applied. The methods of enhancing heat transfer within the adsorber are also
developed, including adding graphite, copper or nickel foams, adsorbent compressed
into blocks or coated on metal tubes, etc. Undoubtedly, these advanced techniques

∗
Corresponding author. Tel.: +86-21-6293-3250; fax: +86-21-6293-2601.
E-mail address: [email protected] (R.Z. Wang).

0960-1481/03/$ - see front matter  2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 6 0 - 1 4 8 1 ( 0 2 ) 0 0 0 4 5 - 9
250 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

Nomenclature
C specific heat (Wkg⫺1K⫺1);
D distance between two adjacent tube centers or diameter of the
evacuated tube (m);
H distance between the center of the tube and the back plate (m);
⌬H adsorption heat (Jkg⫺1);
h heat transfer coefficient (Wm⫺2K⫺1);
I solar intensity (Wm⫺2);
l the length of evacuated tube (m);
m mass (kg);
R, r radius or radial co-ordinate (m);
S area (m2);
T temperature (K);
X, x angle factor, adsorbate content in adsorbent or axial co-ordinate
(m);Greek symbols
r density (kgm⫺3);
a absorptivity;
e emmittance;
t transmittance;
q the angle between beam of radiation and the surface;
s Stefan–Boltzmann constant,of value, 5.67e-8 Wm-2K-4;
l heat transfer coefficient (Wm⫺1K⫺1);
d thickness (m);
f circumferential co-ordinate;subscripts
a ambient;
b back plate;
go outer glass tube;
gi inner glass tube;
l liquid adsorbate;
m metal;
r radiation;
ref reflection;
z adsorbent.

have greatly improved the performance of the system, and therefore speeded up the
commercialization of adsorption refrigeration systems.
Among the research on solid adsorption refrigeration, much attention has been
given to the system powered by solar energy. This is not only because solar energy
is a clear, harmless and renewable resource, but also because exploring and using
solar energy has a profound meaning for a country in the 21st century.
According to the references that can been found up to now, flat plate collectors
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 251

were used in most solar-powered adsorption refrigeration systems [1–5], the average
temperature of desorption is about 100 °C. Headley et al. [6] applied a CPC collector,
and a higher temperature was obtained, but the COP of the system was very low.
Enibe and Iloeje [7] and Antonio Pralon Ferreira Leite [8] have used cylindrical tube
collectors, which have some advantages in sealing and pressure-bearing compared
with flat plate ones, but the heat loss is still large. Bansal et al. [9] investigated
evacuated tubes in solar-powered adsorption systems, but because they were used
to heat air, which in turn was used to heat the adsorber (two heat transfer processes),
the use of solar energy was not efficient. In order to reduce the heat loss in the
adsorber and improve the COP, evacuated tube collectors have been used here.

2. Mathematical models

The evacuated tube used is shown in Fig. 1. The outer and inner tubes are all
glass material and the center mass transfer channel is metallic. In order to reduce
heat losses by radiation, selective material has been coated on the surface of the
inner glass tube. During the daytime, sunlight passes through the outer glass tube
and is absorbed on and heats the inner glass tube, which then heats the adsorbent.

Fig. 1. Evacuated tube adsorber.

252 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

For the analysis and calculation, some assumptions are made:

1. the pressure in the adsorber is homogeneous;

2. the mass resistance and the heat transfer of gaseous refrigerant in the adsorbent
are neglected;
3. the thermal characteristics of the adsorbent and the ambient temperature are con-
stant; and
4. the back plate is an ideal mirror.

2.1. Governing equations for adsorbent

rCp
∂Tz
∂t
∂2Tz
⫽ lz 2 ⫹ lz
∂x
1 ∂Tz
r
r∂r ∂r 冉 冊 1 ∂2Tz
⫹ lz 2 2 ⫹ rz|⌬H|
r ∂j
∂x
∂t
(1)

where ⌬H is adsorption heat, rCP is the total heat capacity, rCP=rzCpz+rlCPlx, x is

the adsorbate content in the adsorbent.

2.2. Governing equation for outer glass tube

∂Tgo ∂2Tgo 1 ∂2Tgo
dgorgoCpgo ⫽ dgolgo 2 ⫹ dgolgo 2 ⫹ lrg(Tgi⫺Tgo) ⫹ lrs(Ts (2)
∂t ∂x Rgo ∂j2
⫺Tgo) ⫹ ha(Ta⫺Tgo) ⫹ Qr ⫹ Iago
where dgo,lgo, ago, Rgo are the thickness, thermal conductivity, absorptance and radius
of outer glass tube, respectively; lrg, lrs are the radiation heat transfer coefficient
between the outer and inner glass tube, outer glass tube and sky, respectively; ha is
the convection heat transfer coefficient between the outer glass tube and air; I is
total solar intensity on the surface of outer tube, including direct, diffuse and reflected
by back plate components I0 and Iref.; Qr is the total radiation heat, including that
between two adjacent tubes and the one between the tube and back plate, that is
Qr=Qgg+Qbg.

2.3. Governing equation for inner glass tube

∂Tgi ∂2Tgi 1 ∂2Tgi
dgirgiCpgi ⫽ dgilgi 2 ⫹ dgilgi 2 ⫹ lrg(Tgo⫺Tgi) ⫹ lgz(Tz (3)
∂t ∂x Rgi ∂j2
⫺Tgi) ⫹ Iagitgo
where, dgi,lgi,,Rg,,agi are the thickness, thermal conductivity, radius and absorptance
of the inner glass tube, respectively; tgo is transmittance of outer tube.

2.4. Governing equation for the center cooling medium channel

∂Tm ∂2Tm 1 ∂2Tm
dmrmCpm ⫽ dmlm 2 ⫹ dmlm 2 ⫹ lmz(Tz⫺Tm) (4)
∂t ∂x Rm ∂j2
where dm,lm,Rm, are the thickness, thermal conductivity and radius of the cooling
medium channel, respectively. lmz is the heat transfer coefficient between the cooling
medium channel and adsorbent.
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 253

2.5. Governing equation for the back plate

∂Tb
mbCpb ⫽ Qrgb ⫹ SI0∂b ⫹ S0hab(Ta⫺Tb) (5a)
∂t
2Rgo
S ⫽ (D⫺ )×l (5b)
sinq
where D, S0, S are the distance and the areas of the back plate and that can be
radiated by sun light between two adjacent tube centers, respectively; l is the length
of vacuum tubes; hab is the convection heat transfer coefficient; Qrgb is the radiation
heat transfer between vacuum tube and back plate; q is solar angle,
q⬎arcsin(2Rgo/D).
Initial condition:
t⫽0 Tz(x,r,j) ⫽ Tgo(x,j) ⫽ Tgi(x,j) ⫽ Tm(x,j) ⫽ Ta (6)
boundary conditions:1. adsorbent
∂Tz
x ⫽ 0, x ⫽ 1 ⫽0
∂r
∂Tz
r ⫽ Rm, lz ⫽ lmz(Tm⫺Tz)
∂r
∂Tz
r ⫽ Rgi lz ⫽ lgz(Tgi⫺Tz)
∂r
Tz(x,r,0) ⫽ Tz(x,r,360°) (7)
2. evacuated tubes and cooling medium channel
∂T
x ⫽ 0, x ⫽ 1 ⫽0
∂r
T(x,r,0) ⫽ T(x,r,360°). (8)

3. Analysis of the radiation heat transfer

3.1. Radiation between outer and inner glass tubes

Because the heat transfer coefficient of the glass tube is small and the position of
its area that can be radiated directly by sunlight changes with the solar angle, the
inhomogeneous temperature field is analyzed here. Fig. 2 shows the radiation area
on the outer glass tubes corresponding to the small area dF1 on the inner glass tube
(arc ABC).
The angle factor X1,2 between two small area dF1 and dF2 is described as follows:
cos(b)cos(g)
XdF2,dF1 ⫽ dF1 (9a)
p × r2
254 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

Fig. 2. The radiation between outer and inner glass tubes.

r ⫽ 冑R2gi ⫹ R2go⫺2RgoRgicos(a) (9b)

b ⫽ arcsin 冉
Rgosina
r 冊 (9c)

g ⫽ b⫺a (9d)
the radiation heat from dF2 to dF1 is
qdF2,dF1 ⫽ sesdF2(T4dF2⫺T4dF1) (10)
where the systematic emmittance es is
(11)
1

冉 冊 冉 冊
es ⫽
1 1 dF2 1
⫺1 ⫹ ⫹ ⫺1
ego XdF2,dF1 dF1 egi
where, s is the Stefan Boltzmann constant, of value 5.67×10⫺8Wm⫺2K⫺4; egi and
ego are the emittance of the inner and outer tubes, respectively.

3.2. Radiation heat transfer between two adjacent tubes

Generally because of the limit to the mass of adsorbent in an evacuated tube, a

number of evacuated tubes are needed to meet the given cooling capacity. They are
usually assembled in parallel, thus radiation heat transfer will occur between two
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 255

adjacent tubes. As shown in Fig. 3, the small area dF1 on tube 1 has a range of
radiation (arc BFC) on tube 2.
The angle factor is calculated as follows:
cosb × cosg
XdF1,dF2 ⫽ dF2 (12)
p × r2
l21 ⫽ D2 ⫹ R2go ⫹ 2DRgocosq (12a)
l22 ⫽ D2 ⫹ R2go⫺2DRgocosa (12b)

f ⫽ arccos 冉l21 ⫹ D2⫺R2go

2l1D 冊 (12c)

r2 ⫽ R2go ⫹ l21⫺2Rgol1cos(a⫺f) (12d)

l21⫺(r2 ⫹ R2go)
cosb ⫽ (12e)
2Rgor
l22⫺(r2 ⫹ R2go)
cosg ⫽ (12f)
2Rgor

Fig. 3. Radiation between two adjacent tubes.

256 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

qgg ⫽ sesdF2(T4dF1⫺T4dF2). (13a)

3.3. Radiation between evacuated tube and back plate

The radiation area BC of back plate corresponding to dF1 on the evacuated tube
is shown in Fig. 4(a), and areas AEB and CFD on the tube to back plate in Fig.
4(b). The calculation of BC is described by Eq. (14a,b,c,d,e):
dF1O22 ⫽ R2go ⫹ D2⫺2RgoDcosg (14a)
dF1E2 ⫽ dF1O22⫺O2E2 ⫽ D2⫺2RgoDcosg (14b)
dF1E
O2C2 ⫽ dF1O22 ⫹ (dF1E ⫹ EC)2⫺2dF1O2 × (dF1E ⫹ EC) × (14c)
dF1O2
EC2 ⫽ O2C2⫺R2go dF1D2 ⫽ Rgosing ⫹ H (14d)
BC ⫽ 冑dF1C2⫺dF1D2⫺(Rgosing ⫹ H) × tgg (14e)
angle factor XdF1,p:

Fig. 4. Radiation between evacuated tube and the back plate.

 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 257

cosacosb
XdF1,P ⫽ F (15a)
p × r2 p
DB ⫽ dF1D × tgg (15b)

DP ⫽ DB ⫹ BC / 2
r2 ⫽ dF1D2 ⫹ DP2 (15c)

dF1D ⫽ Rgosing ⫹ H
dF1D
cosb ⫽ (15d)
r

dF1D
a ⫽ g ⫹ arcsin( ).
r

3.4. Reflection of back plate

Using a back plate (reflective plate) can increase the efficiency of sunlight collec-
tion and improve the performance of adsorber. As to the given angle q, the reflection
of sunlight on the evacuated tube by the back plate is shown in Fig. 5. It is not
difficult to know that there exit the relationship a=p/2+q (a) or a=p+g (b). By calcu-
lating the values of b and g by using Eqs. (16a,b) and (17a,b), the reflection area
ACB onto the evacuated tube can be defined.
O2F ⫽ O2D⫺DE⫺EF (16a)
thus
Rgo
⫺2H × ctgq ⫹ Rgo × ctgqcosb ⫽ Rgosinb (16b)
sinq
O1D ⫹ DE ⫹ EF ⫹ FO2 ⫽ D (17a)
thus
Rgo
⫹ 2H × ctgq⫺Rgo × ctgqsing ⫹ Rgocosg ⫽ D (17b)
sinq
.

4. Numerical calculation and analysis

According to Eqs. (1)–(17), numerical calculations have been done with a zeolite–
water pair for the evacuated tube adsorber. Some parameters are: ambient tempera-
258 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

Fig. 5. Reflection of the back plate.

ture Ta=30 °C, cooling temperature Tc=35 °C, evaporation temperature Te=10 °C;
ego=0.9, ago=0.05, tgo=0.91, agi=0.95; H=Rgo+1 mm, Dgo=50, 60, 70, 80, 90, 100,
110 mm, Dgi=Dgo⫺10 mm, the total solar intensity I0 is shown in Fig. 6. A Dubinin–
Astakhov equation is used in this work:
ln{x / (w0r)} ⫽ ⫺D[Tln(P / P0)]n (18)
where w0 is the total volume of the micropores accessible to the vapor; x is the mass
of adsorbate condensed in the micropores of the adsorbent at temperature T and
relative pressure P/P0, P0 is the saturated pressure corresponding to the evaporating
temperature. D and n depend on the pair; r is the specific mass of the adsorbate.
The values of w0, D, n are given in Ref. [10]: 0.269 m3kg⫺1, 1.80e-7,2.

4.1. The temperature distribution in the evacuated tube adsorber

In order to clearly depict the temperature distribution in the adsorber, Fig. 7 gives
the grids for calculation. The numerical result is shown in Fig. 8, where the time of
radiation of sunlight is from 6:00 to 14:00 h, and the diameter Dgo of the evacuated
tube is 70 mm, the distance between two adjacent tube centers is 2Dgo. Fig. 8(a) is
the picture at 10:00 h in the morning and Fig. 8(b) gives the picture at 14:00 h. In
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 259

Fig. 6. Solar intensity changing with time.

Fig. 7. Grids on circumferential direction.

(a) the temperature gradients are very large both in the radial direction and between
the up part and down part, while in Fig. 8(b) it is less in circumferential direction.
The main reason is that the heat transfer coefficient of the adsorbent is small (here
equal to 0.102 Wm⫺1K⫺1) and the direction of sunlight changes with time.

4.2. The effect of the diameter of the evacuated tube on its performance

When the solar intensity is given, the temperature of the evacuated tube collector
is usually limited by the quality of adsorbent, in other words the diameter will have
an effect on the performance and cooling capacity of an evacuated tube adsorber.
Simulation results of seven evacuated tubes with different diameters have been indi-
260 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

Fig. 8. The temperature distribution in evacuated tube adsorber. (a) at 10:00; (b) at 14:00.

cated in Fig. 9, where the distance between two adjacent tube centers is two times
its diameter and the diameter of the cooling medium channel is 20 mm. The curves
in Fig. 9(a) show the relationships between the diameter and the minimum, average
and maximum temperatures. The larger the diameter, the less the minimum and
average temperatures, but the maximum temperature begins to tend to be a constant
at Dg=70 mm. The relationships betwewen the diameter and performance are given
by the curves in Fig. 9(b); when the diameter is equal to 70 mm, both the COP and
the cooling capacity reach their maximum values 0.255 and 4377 kJm⫺2, respect-
ively. By comparing the curves in Fig. 9(a) and (b), we find that although the average
temperatures for the tubes whose diameters are less than 70 mm can show higher
values, their COP and cooling capacity may be lower due to the smaller mass quan-
tity of the adsorbent.
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 261

Fig. 9. Performances vs. diameter of evacuated tube, (the distance of two adjacent tube centers D=2Dgo,
Db = 20mm).

4.3. The effect of distance between two adjacent tube centers on its performance

In order to reduce the total area of the adsorber and its manufacturing costs, the
effect of the distance between two adjacent tubes on its performance is also analyzed.
Fig. 10 indicates the changes of average temperature, COP and cooling capacity
with different distances. It is known from Fig. 10(b) that the cooling capacity is
sensitive to the distance and is greatest at about two-times the distance; nevertheless,
the maximal average temperature in Fig. 10(b) has great sensitivity only when the
distance is less than 2.5-times the diameter, then it tends to become a constant. As
to the curve of cooling capacity, the reason may be that, on one hand, when the
distance is less than two-times the diameter, the average temperature is lower, so is
262 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

Fig. 10. The relationship of the distance and its performance, (Ddo = 70mm, Db = 20mm).

the cooling capacity; on the other hand, when the distance is larger than that value,
the total quality of adsorbent will be less, hence, the cooling capacity will also not
be satisfied.

4.4. The effect of mass quantity of adsorbent on its performance

Fig. 11 has shown the relationships of the diameter of the cooling medium channel
or mass quantity of adsorbent in an evacuated tube on the performance. From Fig.
11(a) and (b), it is known that for a given diameter of an evacuated tube, the best
performances are under conditions when the diameter of the cooling medium channel
Db is about equal to Dgo minus 40 mm, which means the optimal thickness of adsorb-
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 263

Fig. 11. The relationship of performance and diameter of cooling medium channel, (a) COP vs. Db; (b)
cooling capacity vs. Db.

ent in an tube is 15 mm. Also, it indicates that the greater the diameter of an evacu-
ated tube, the better the performance.

4.5. The economic analysis

When assessing the performance of a refrigeration system or heat pump, its COP
value is usually considered, which undoubtedly can account for the economic charac-
ter well in the system powered by electricity or oil-boiling. But, if it is powered by
solar energy, more attention will be paid to the ratio of the cooling capacity to the
costs of materials and manufacture. Therefore, another assessing index is used here,
and according to it the optimum structure of the system is determined. The defined
264 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

parameter is Solar Powered Adsorption (SPA), SPA=Qf/(C×A), where Qf is the coo-

ling capacity, C is the total cost per unit area and A is the area of the adsorber.
Obviously, a larger parameter SPA may has a better economic character and perform-
ance.
Tables 1 and Table 2 show the numerical simulation results of evacuated tube
adsorbers with different diameters (where D=2Dgo) and those with different distances
between two adjacent tube centers (where Dgo=70 mm), respectively. Fig. 12 depicts
the corresponding curves of the assessing parameter SPA changes. The curves in
Fig. 12(a) indicate that when applying the above assessing SPA, using evacuated
tubes with a larger diameter may be more economic, and the optimal thickness of
adsorbent is about 10 mm less than 15 mm shown by Fig. 11. Fig. 12(b) shows us
that the optimal distance between two adjacent tube centers is equal to about 2.5-
times the diameter of outer tube.
The above calculations are based upon the thermal conductivity of zeolite (0.102
Wm⫺1K⫺1), however if a compressed adsorbent bed or solidified bed with incorpor-
ation of metals was used the thermal conductivity would be improved greatly. For
example, a suitable zeolite bed can have a thermal conductivity of 1 Wm⫺1K⫺1: in
this case the cooling capacity and parameter SPA both increase greatly as shown in
Table 3 and Fig. 13. Fig. 13(b) also shows us that a larger increment of cooling
capacity will be obtained, with more adsorbent in the evacuated tube, when the heat
transfer coefficient of adsorbent increases from 0.102 to 1 Wm⫺1K⫺1. This is because
the temperature gradient in the adsorbent bed has decreased and consequently more
adsorbate is desorbed.

5. Summary

Through numerical simulation and analysis on the evacuated tube adsorber, it is

found that structure parameters, such as the diameter of the evacuated tubes, the
mass quantity of adsorbent and the distance between two adjacent tube centers, have
important effects on the maximum average temperature and the performance of the
system. It seems to indicate that evacuated tubes with a larger diameter will be better.
The results of two different heat transfer coefficients of adsorbent zeolite, 0.102 and
1 Wm⫺1K⫺1, have both shown the optimum thickness of adsorbent is about 10 mm,
whose parameter SPAs are 58.81 and 61.97, respectively. But the optimal thickness
of adsorbent is about 15 mm when the largest COP is considered.

Acknowledgements

This work was supported by China Key Fundamental Research Program under
the contract No. G2000026309.
Table 1
The performance of evacuated tube adsorber with different diameters(D=2×Dgo,lz=0.102 Wm⫺1K⫺1)

Diameter Diameter Evacuated tube Zeolite Area Cost of Cooling Total cost Parameter
Dgo (mm) Db (mm) (m2) manufacture capacity (US$/m2) SPA
(US$/number) (kJ/m2)

(number) price mass price

(US$) (kg) (US$)

70 20 8 36 16.08 49.4 1.05 2.5 4377.02 99.3 44.08

30 13.57 40.7 2.8 4607.36 94.0 49.01
40 10.05 30.2 3 4458.09 85.8 51.96
50 5.53 16.6 3.3 3223.73 74.8 43.10
80 20 7 35 19.79 59.4 1.04 2.5 4315.5 107.5 40.14
30 17.59 52.8 2.8 4729.15 102.9 46.00
40 14.51 43.5 3 4930.78 95.7 51.52
50 10.56 31.7 3.3 4745.63 86.0 55.18
60 5.72 17.2 3.5 3390.23 73.7 46.00
90 20 7 39.4 26.39 79.2 1.17 2.5 4054.44 116.3 34.86
30 24.19 72.5 2.8 4583.88 112.1 40.89
40 21.11 63.3 3 4992.28 105.5 47.32
50 17.15 51.4 3.3 5146.88 96.9 53.12
60 12.32 36.9 3.5 4934.43 85.9 57.44
70 6.60 19.8 3.8 3495.90 72.8 48.02
100 20 6 37.5 29.03 87.1 1.1 2.5 3685.85 126.9 29.05
C.H. Li et al. / Renewable Energy 28 (2003) 249–269

30 27.14 81.4 2.8 4314.47 123.1 35.05

40 24.50 73.5 3.0 4857.09 117.3 41.41
50 21.11 63.3 3.3 5199.16 109.4 47.52
60 16.96 50.9 3.5 5379.54 99.4 54.12
70 12.06 36.2 3.8 5139.60 87.4 58.81
80 6.41 19.2 4.0 3626.70 73.4 49.41
265
Table 2 266
The performance of evacuated tube adsorber with different distance(Dgo=70 mm, lz=0.102 Wm⫺1K⫺1)

Distance Vacuum Zeolite Area (m2) Cost of Cooling Total cost Parameter
between tube manufacture capacity (US$/m2) SPA
two tubes (US$/number) (kJ/m2)
(×Dg)

(number) Price Mass (kg) Price

(US$) (US$)

1.25 12 54.0 24.13 72.4 1.033 2.5 2902.27 151.4 19.17

1.5 10 45.0 22.12 66.4 1.015 2.5 3424.12 136.8 25.03
2 8 36.0 16.09 48.3 1.050 2.5 4377.02 99.3 44.08
2.5 7 31.5 14.08 42.2 1.120 2.5 4075.87 81.4 50.07
3 6 27.0 12.07 36.2 1.12 2.5 3510.7 69.8 50.30
C.H. Li et al. / Renewable Energy 28 (2003) 249–269
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 267

Fig. 12. The curves of assessing parameter SPA, where lz=0.102 Wm⫺1K⫺1.
268 C.H. Li et al. / Renewable Energy 28 (2003) 249–269

Table 3
The performance of evacuated tube adsorber with different diameters(D=2×Dgo,lz=1 Wm⫺1K⫺1)

Diameter Diameter Db COP Cooling capacity Parameter SPA Increment of

Dgo (mm) (mm) (kJm⫺2) cooling capacity
(%)

70 20 0.285 4893.15 49.29 11.79

30 0.293 5022.09 53.42 9.01
40 0.276 4737.06 55.17 6.26
50 0.189 3248.84 43.41 0.78
80 20 0.289 4962.50 46.13 14.99
30 0.308 5295.02 51.46 11.97
40 0.312 5356.96 55.98 8.64
50 0.292 5021.66 58.40 5.82
60 0.199 3414.72 46.33 0.72
90 20 0.281 4818.35 41.44 18.84
30 0.307 5268.95 46.99 14.95
40 0.323 5542.79 52.53 11.03
50 0.325 5575.52 57.56 8.33
60 0.303 5208.67 60.60 5.56
70 0.205 3523.63 48.40 0.79
100 20 0.264 4535.78 35.74 23.06
30 0.295 5062.34 41.12 17.33
40 0.319 5472.38 46.66 12.67
50 0.337 5789.06 52.92 11.35
60 0.338 5802.21 58.35 7.86
70 0.316 5418.74 61.97 5.43
80 0.212 3647.42 49.70 0.57

References

[1] Wang RZ. Adsorption refrigeration research in Shanghai Jiao Tong University. Renew Sustainable
Energy Rev, in press. 2000.
[2] Pons M. Experimental data on a Solar-powered ice maker using activated carbon and methanol
adsorption pair. Trans ASME J Sol Energy Eng 1987;109(4):303–10.
[3] Passos EF, Escobedo SF, Meunier F. Simulation of intermitent adsorption Solar Cooling System.
Sol Energy 1989;42(2):103–11.
[4] Hajji A, Worek WM, Lavan Z. Dynamic analysis of a closed-cycle solar adsorption refrigerator
using two adsorbent–adsorbate pairs. J Sol Energy Eng 1991;113:73–9.
[5] Critoph RE. Ammonia carbon solar refrigerator for vaccine cooling. Renew Energy 1994;5(1-
4):502–8.
[6] Headley OSte, Kohdinala AF, Doom IA. Charcoal–methanol adsorption refrigerator powered by a
compound parabolic concentrating solar collect. Sol Energy 1994;53(2):191–7.
[7] Enibe SO, Iloeje OC. Design optimization of the flat plate collector for a solid adsorption solar
refrigerator. Sol Energy 1997;61(2):77–87.
 C.H. Li et al. / Renewable Energy 28 (2003) 249–269 269

Fig. 13. The changes of parameter SPA and increment of cooling capacity with the diameter of cooling
medium channel (or mass quantity of adsorbent) where lz=1 Wm⫺1K⫺1.

[8] Leite Antonio Pralon Ferreira, Daguenet Michel. Performance of a new solid adsorption ice maker
with solar energy regeneration. Energy convers Manage 2000;41:1625–47.
[9] Bansal NK. Performance testing and evaluation of solid absorption solar cooling unit, Sol Energy,
61(2), 127-140. Sol Energy 1997;61(2):127–40.
[10] Meunie, F, Douss N. Performance of adsorption heat pumps: active carbon–methanol and zeolite–
water pairs. ASHRAE Trans. 1990.