U

n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
Performance improvement of adsorption cooling by
heat and mass recovery operation
R.Z. Wang *
Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200030, China
Received 15 February 2000; received in revised form 19 September 2000; accepted 16 December 2000
Abstract
An adsorption cooling system was developed and tested and various operation procedures have been tried. The
experimental results show that the heat recovery operation between two adsorption beds will increase the COP by
about 25% if compared with one adsorber basic cycle system. It was also proved that mass recovery is very eective for
heat recovery adsorption cooling operation, which may help to obtain a COP increase of more than 10%. Theoretical
analyses on the COP have been completed for various heat and mass recovery cycles, such as basic intermittent
adsorption cycle, continuous two-adsorber heat recovery cycle, mass recovery cycle, mass recovery with sensible heat
recovery, and mass recovery with both sensible heat and heat of adsorption recovery. The theoretical results are in
good agreement with experimental values. Based upon the developed theoretical model, it is possible to predict the
COP for various operation procedures of a real adsorption cooling system. # 2001 Published by Elsevier Science Ltd
and IIR. All rights reserved.
Keywords: Adsorption; Cooling; Air conditioning; Heat recovery; Mass recovery
0140-7007/01/$20.00 # 2001 Published by Elsevier Science Ltd and IIR. All rights reserved.
PI I : S0140- 7007( 01) 00004- 4
International Journal of Refrigeration & (&&&&) &&
www.elsevier.com/locate/ijrefrig
JIJR 513p Disk used DTD=4.1.0
* Tel./fax: +86-21-6293-3250.
E-mail address: rzwang@mail.sjtu.edu.cn
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
1. Introduction
Adsorption refrigeration with dierent cycles have
been studied extensively, the typical adsorption refrig-
eration cycles are:
1. basic cycle, which is an intermittent refrigeration
cycle if operated with one bed. Continuous refrig-
eration can be only achieved with two or more
beds to shift the heating-cooling phases. In a basic
cycle, there is no heat recovery;
2. continuous heat recovery cycle, which is usually
operated with two adsorption beds. The adsorber
to be cooled will transfer its heat to the adsorber
to be heated, the heat transfer between two beds
includes sensible heat and heat of adsorption.
This heat recovery process is useful in increasing
COP. Multi-beds could be also adopted to get
more heat recovery and thereby much higher
COP;
3. thermal wave cycle [1], in which it is assumed that
a big temperature gradient exists along an
adsorption bed. For a two bed system, high tem-
perature thermal uid ows into the adsorber,
exchanges heat with the bed, and the temperature
goes down along the bed rapidly, thus the outlet
temperature will be close to ambient. After being
cooled by ambient surroundings, the uid ows
into another adsorption bed, absorbes heat from
the bed, and the temperature of the uid goes up.
At the exit of this bed, the thermal uid tempera-
ture will be very close to the temperature of heat
source. In this case, only less heat is added to the
system, and less heat released to the environment,
thus heat recovery ratio is high and the COP is of
course signicantly increased;
4. convective thermal wave cycle [2], the concept is
the same as thermal wave cycle, however the
thermal uid for heating or cooling to the beds is
initiated by the refrigerant itself, thus the heat
transfer between thermal uid and adsoption bed
is direct contact heat transfer, which is incorpo-
rated with mass transfer in the system;
5. cascading cycle [3,4], in which zeolitewater/acti-
vated carbon-methanol, or, zeolitewater/silica gel
water, and zeolitewater/zeolitewater are usually
used for cascading, the high temperature heat
source (e.g. 200

C) is used to drive the high tem-

perature stage adsorption refrigeration cycle
(typically 100200

C for zeolitewater). The low

temperature stage adsorption refrigeration is dri-
ven by the sensible heat and heat of adsorption
of high temperature stage, for example, activated
carbonmethanol or silica gelwater adsorption
refrigeration cycles are suitable for generation
temperature of 100

C, which are operated

between 30100

C.
Nomenclature
B
0
adsorption parameter (K
2
)
C
pa
specic heat of adsorbent (kJ/kg K)
C
pr
specic heat of refrigerant liquid (kJ/kg
K)
COP refrigeration COP
f fugacity (Pa)
L latent heat of evaporation of refrigerant
(kJ/kg)
M
a
mass of adsorbent (kg)
P
c
condensing pressure (Pa)
P
e
evaporation pressure (Pa)
Q
g
heat for isobaric generation process (kJ)
Q
h
heat for isometric heating process (kJ)
Q
ref
refrigeration eect (kJ)
Q
reg
heat regenerated (kJ)
R heat capacity ratio of adsorber material
and thermal uid to adsorbent
R
f
heat capacity ratio of adsorber material
to adsorbent
R
m
heat capacity ratio of thermal uid to
adsorbent
SCP specic cooling power (kw/kg adsor-
bent)
T temperature (

C, K)
T
a1
temperature to start adsorption (

C)
T
a2
adsorption temperature (

C)
T
b
adsorbent bed temperature (

C)
T
c
condensing temperature (

C)
T
e
evaporation temperature (

C)
T
g1
temperature to start desorption (

C)
T
g2
desorption temperature (

C)
T
s
saturated temperature (

C)
T
i
,T
reg
temperature after heat recovery (

C)
X, X
1
, X
2
adsorption capacity (kg refrigerant/kg
adsorbent)
x
0
adsorption capacity at a saturated pres-
sure p
s
corresponding to T
s
(kg/kg)
half width of Guess normal distribution
x adsorption capacity dierence between
adsorption phase and desorption phase
x = x
1
x
2
(kg/kg)
H heat of adsorption or desorption (kJ/kg)
dx enlarged desorption concentration
changes due to mass recovery (kg/kg)
2 R.Z. Wang / International Journal of Refrigeration &(&&&&) &&
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
The above adsorption refrigeration cycles have been
investigated by various researchers. The basic cycle is
easy to apply and is very suitable for solar refrigeration
and also for waste heat recovery refrigeration. The heat
recovery cycle is practical as the system has reasonable
COP and is not so complicated. The thermal wave cycle
is attractive; however, it needs very high thermal con-
ductivity which is not easy for adsorbents to reach. In
addition, the thermal uid velocity is restricted to meet
the demand, which reduces the specic cooling power
[5]. The convective thermal wave cycle is reasonable in
theory but is not so convenient to build the real system.
Cascading cycle is attractive to have high COP, but the
system is very complicated [4].
For real applications, the heat recovery two bed sys-
tem is well accepted, however one needs to increase its
COP and specic cooling power (SCP) signicantly.
Obviously the various operating procedures will inuence
the system performance, particularly the mass recovery
operation [6], which consists of connecting the two beds,
equalizing their pressure before heat recovery. This pro-
cess will enlarge the desorbed refrigerant mass, and
thereby increase the refrigeration eect, and possibly
have an increased COP. This paper shows the detailed
mass recovery theory and its applications in experiments
of a two bed adsorption air conditioning system.
2. Heat recovery adsorption refrigeration cycles
Adsorption refrigeration is based upon the processes
of heatingdesorptioncondensation and cooling
adsorptionevaporation, the coolingadsorption process
needs heat dissipation both of sensible heat and heat of
adsorption. A two bed continuous adsorption refrigera-
tion system with heat recovery is shown in Fig. 1. The
schematic shows the operation when adsorber 1 is
cooled and connected to the evaporator to obtain
adsorption refrigeration in the evaporator, and adsorber
2 is heated and is connected to the condenser to obtain
heatingdesorptioncondensation. The condensed
refrigerant liquid ows into the evaporator via a ow
control valve. The operation phase can be changed, and
the go-between will be a short time heat recovery pro-
cess in which the two pumps drive the thermal uid in
the circuit between two adsorbers (the connection to the
heater and cooler are blocked during this process). Heat
recovery is important to increase the cycle COP, the
possible heat recovery for a two adsorption bed system
will be some part of sensible heat and heat of adsorption
shown as Fig. 2.
The COP for basic type adsorption refrigeration cycle
(one bed or two beds without heat recovery) is
COP
B
=
Q
ref
Q
h
Q
g
(1)
The COP for a two-bed heat recovery cycle can be
expressed as
COP =
Q
ref
Q
h
Q
g
Q
reg
(2)
where Q
reg
is the heat recovered. In the above equations,
Q
h
and Q
g
are the heat for generation corresponding the
two processes (isosteric and isobaric). The refrigeration
eect
Q
ref
= M
a
xL M
a
x

T
c
T
e
C
pl
dT (3)
which is the latent heat L multiplied by the cycle refrig-
erant. However, some of the cooling quantity will be
consumed to cool the refrigerant liquid from condensing
temperature T
c
to evaporation temperatureT
e
.
The detailed calculation method has been shown in
earlier work [7]. The calculation in this paper is based
Fig. 2. Two-bed adsorption system with heat recovery.
Fig. 2. Two-bed adsorption system with heat recovery.
Fig. 1. Schematic of a two-bed adsorption refrigeration system
with heat recovery.
Fig. 1. Schematic of a two-bed adsorption refrigeration system
with heat recovery.
R.Z. Wang / International Journal of Refrigeration &(&&&&) && 3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
upon adsorption refrigeration pair, activated carbon
methanol, in which Shanghai YK activated carbon
made from coconut shell is used. The fundamental
adsorption data can be found in the literature [8,9].
Typical adsorption characteristics are: rated surface
area 1200 (m
3
/g), density 0.62 kg/l and size 820 (mesh).
The adsorption equation can be described by a modied
adsorption equation [8]
x = x
0
exp B
0
y ( ) exp y
2

2
=2

1 erf z ( ) [ ]=2 (4)
where the variable y = T= ( )
2
ln
2
f
0
=f ( ) contains the
adsorptive-dependent shifting factor anity factor),
and therefore simplies the overall graphical repre-
sentation, f
0
and f represent fugacity of refrigerant at
saturated pressure and adsorption pressure respectively.
The total adsorption concentration z = y B
0
=
2

=

2
_
, and erf(z) is error function. This equation con-
tains three parameters:x
0
, B
0
and, which can be
determined by adsorption experiments. The measured
isobars and isotherms showed the typical value of
YKAC-methanol adsorption is x
0
=0.294 kg/kg, B
0
=
1.03310
6
K
2
, =0.28910
6
K
2
.
As shown in Fig. 3, the cycle COP will be increased
more than 25% by heat recovery process, but the COP
for a single eect refrigeration system is still low, possi-
bly in the range of 0.4 to 0.7. Multi-beds system will
recover more heat, thereby increasing the COP. How-
ever, the real system will be very complicated. The
modeling of the heat recovery process has assumed a
uniform temperature bed, the adsorber conguration
has been added in the text. The experimental results for
the two bed temperatures after heat recovery were
normally about 5 to 10

C however, for simplicity, the

theoretical consideration has assumed non temperature
dierence between two beds after heat recovery.
The heat recovery may include sensible heat recovery
and adsorption heat recovery. Regarding the dierent
heat recovery (sensible heat and adsorption heat recov-
ery), it is easily explained by the pTx diagram for the
adsorption system. If the heat recovery temperature T
reg
is bigger than T
a1
, no adsorption heat can be recovered,
but if T
reg
is less than T
a1
, the heat of adsorption at the
evaporation pressure and in the temperature range [T
reg
,
T
a1
] exists and thereby can be possibly recovered in
addition to the sensible heat recovery. The theoretical
work to simulate heat recovery cycle was also shown in
the early publication [7]. A separating point for sensible
heat recovery or sensible heat and adsorption heat
recovery happens when the desorption temperature T
g2
is large enough which causes T
a1
>T
reg
.
Whether the heat of adsorption can be recovered is
dependent on the valve operation between the two beds
during the heat recovery process, when the bed tem-
perature is less than T
a1
and if the adsorber is not con-
nected to the evaporator, no heat of adsorption
happened and thus no adsorption heat recovery. But if
the adsorber is connected to the evaporator, the heat of
adsorption will thereby be generated in the adsorber bed
and thus will be possible to transfer to the bed of gen-
eration for recovery.
For a real heat recovery adsorption system, the heat
capacity of the metallic adsorber and also the thermal uid
will have a strong inuence on the COP system. If R
m
is
dened as the heat capacity ratio of adsorber material to
adsorbent, and R
f
that of thermal uid to adsorbent, the
total heat capacity ratio R=R
m
+R
f
will have a strong
inuence on the COP system. Atypical example of the heat
capacity ratio eect on system COP is shown in Fig. 4. It
Fig. 3. COP with and without heat recovery (T
e
=5

C,
T
a
=T
c
=30

C), 1-basic type cycle, 2-heat recovery cycle (*,

sensible heat recovery, &, sensible+adsorption heat recovery).
Fig. 3. COP with and without heat recovery (T
e
=5

C,
T
a
=T
c
=30

C), 1-basic type cycle, 2-heat recovery cycle (*,

sensible heat recovery, &, sensible+adsorption heat recovery).
Fig. 4. Heat capacity ratio inuence on COP of a heat recovery
system. No consideration of the heat capacity of adsorber
material and thermal uid, R=0; 2, R=1.85; 3, R=2.9.
Fig. 4. Heat capacity ratio inuence on COP of a heat recovery
system.No consideration of the heat capacity of adsorber mate-
rial and thermal uid, R=0; 2, R=1.85; 3, R=2.9.
4 R.Z. Wang / International Journal of Refrigeration &(&&&&) &&
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
is obvious that the COP decreases signicantly if the
ratio R increases. For example, when R=2.9, the COP
will decrease by 40% if compared with the ideal COP
0
(COP corresponding to R=0) . When R=1.85, the COP
decreases about 2530%. The above two examples are
related to the real design of a prototype adsorption air
conditioning system, in which R=2.9 and R=1.85 are
corresponding to water or oil as the thermal uid
respectively [10,11].
In the real design of an adsorption system good heat
transfer should be considered in order to shorten the
cycle time and increase specic cooling power (SCP),
which may need to increase heat transfer area by nned
tubes, etc. However, the heat capacity ratio should be
controlled, the ideal value of R should be lower than 5,
possibly below 3. Oil should be used as thermal uid to
decrease the heat capacity ratio R
f
, also the ow volume
of thermal uid should be controlled.
As shown in Fig. 5, a novel adsorber has been
designed which is applied in activated carbon-methanol
adsorption air-conditioning system. The newly designed
plate-nned tubes adsorber is successful both for heat
transfer and also the heat capacity control. The plate-
nned tubes with a diameter of 9.5 mm are used for heat
transfer between the thermal uid and the adsorption
bed, the aluminum plates are incorporated with the
tubes to extend heat transfer surface. Adsorbent bed is
packed by metallic net with the heat exchanger tubes,
and a space to the shell is kept to allow gas ow chan-
nels for enhancing mass transfer. Inside the adsorber an
extra mass ow channel is designed to separate the bed
into two halves. The adsorber heat capacity ratio is
R
m
=1.36, while the thermal uid heat capacity ratios
are R
f
=1.5 for water and R
f
=0.49 for oil. The total
heat capacity ratio is thereby controlled as R=1.85 if oil
is used as the thermal uid.
3. Heat and mass recovery adsorption refrigeration cycle
A normal two bed heat recovery system can be
demonstrated in Fig. 6 in a pTx diagram (if compared
with Fig. 2). Mass recovery could be initiated before
heat recovery (the ideal heat recovery state will be ee
/
for a two bed system shown in Fig. 6). As is clear, when
adsorber 1 (as generator) is desorbed, it is at the gen-
eration temperature T
g2
and condensing pressure P
c
,
which is to be cooled to serve as adsorber (temperature
from T
g2
to T
a2
, pressure from P
c
to P
e
), while adsorber
2 (currently as adsorber) has adsorbed refrigerant, and
is hoped to be heated to serve as generator (temperature
from T
a2
to T
g2
, pressure from P
e
to P
c
).
More desorption could be achieved by pressure
reduction of adsorber 1, or more adsorption could be
obtained if the pressure of adsorber 2 is increased. In a
closed adsorption system, the connection between two
adsorbers will meet such demands, the two beds pres-
sure will be equalized by the internal mass recovery, the
two bed pressure will reach an equilibrium pressure
P
m
- (P
e
+P
c
)/2. This process will cause more deso-
rption in the generator shown as Fig. 7. As is shown in
Fig. 8, the operation for a two bed mass recovery can be
Fig. 5. A newly designed plate-nned tubes adsorber.
Fig. 5. A newly designed plate-nned tubes adsorber.
Fig. 6. Diagram of the intermittent and heat recovery cycle.
Fig. 6. Diagram of the intermittent and heat recovery cycle.
Fig. 7. Diagram of mass recovery cycle and heat and mass
recovery cycle.
Fig. 7. Diagram of mass recovery cycle and heat and mass
recovery cycle.
R.Z. Wang / International Journal of Refrigeration &(&&&&) && 5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
easily achieved by the connection of adsorber 1 and
adsorber 2 via a valve. The mass recovery cycle (a
2
a
3

g
1
/
g
1
g
2
g
3
a
1
/
a
2
) extends the two bed basic cycle or
two bed heat recovery cycle (a
2
g
1
g
2
a
1
a
2
showns as
Fig. 7), and the cycle mass is increased from x to
x+dx, which causes the refrigeration eect to
increase. If the heat of adsorption and the heat of deso-
rption is the same (at the same pressure), the cycle COP
will increase due to the increased refrigeration eect in a
cycle.
In this discussion the mass recovery process is
assumed to be adiabatic.The process is divided into two
parts operating in two beds as a
2
a
3
and g
2
g
3
. The
process can be simulated and calculated utilizing the
following model.
The vapor desorbed from the high pressure bed is
entirely re-adsorbed by the low pressure bed. That is
X
a3a2
= X
g3g2
(4)
The beds are adiabatic during this process. The tem-
perature variation is caused by sorption or desorption.
(C
pa
X
1
C
pr
) (T
a3
T
a2
) = HX
a3a2
(5)
(C
pa
X
2
C
pr
) (T
g3
T
g2
) = HX
g3-g2
(6)
where C
p
is the specic heat of refrigerant liquid, X
1
and
X
2
are the adsorptivity at the concentrated adsorbed
state and diluted desorbed state, respectively, H is the
heat of adsorption/desorption. The subscripts `a' and `r'
represent adsorbent and refrigerant, respectively. Here
the void volume is neglected in this model, and the heat
of adsorption/desorption is assumed as a constant at the
corresponding averaged pressure.
The nal pressure of the two beds should be equal to
each other,
P
g3
= P
a3
(7)
The iterative calculation method could be used by
supposing a nal pressure at the rst P=(P
c
+P
e
)/2,
and calculate the temperature and concentration change
of each bed. Then modify the supposed pressure if
necessary to satisfy Eq. (4) until a satisfactory result is
derived.
The typical comparison of COP between basic type
cycle (two beds but without heat recovery) and mass
recovery cycle (two beds without heat recovery) is
shown in Fig. 9. It is seen that mass recovery cycle is
practically suitable for low generation temperatures, the
increased COP is in the range of 10100%.
The mass recovery process is usually used before the
heat recovery process, the combined mass and heat
recovery procedures will increase COP signicantly. In
the real system operation, there may be two kinds of
heat recovery. One of which is sensible heat recovery,
the other is sensible heat recovery followed by adsorp-
tion heat recovery. Fig. 10 shows the ideal COP (for
R=0) of various operation procedures compared with
basic type cycle and mass recovery cycle. It is seen that
mass recovery followed by heat recovery (sensible and
heat of adsorption) gives the best performance, and for
activated carbonmethanol air-conditioning system,
COP over 0.6 can be achieved with a generation tem-
perature of 80

C. If the generation temperature reaches

120

C, COP will be close to 0.8.

Realistic calculated results for the COPs of mass and
heat recovery cycle can be shown in Fig. 11, in which
two heat capacity ratios R=2.9 and R=1.85 have been
Fig. 9. Comparison of COP for basic type cycle (1) and mass
recovery cycle (2). T
e
=5

C, T
c
=T
a
=30

C.
Fig. 9. Comparison of COP for basic type cycle (1) and mass
recovery cycle (2). T
e
=5

C,T
c
=T
a
=30

C.
Fig. 8. Mass recovery operation between two adsorbers via
opening of valve A.
Fig. 8. Mass recovery operation between two adsorbers via
opening of valve A.
6 R.Z. Wang / International Journal of Refrigeration &(&&&&) &&
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
assumed. It is seen that the reasonable COP for a
real adsorption air conditioning system can be close
to 0.6.
4. Performances of an adsorption air conditioning
system
An adsorption air conditioning system using plate n
heat exchangers as adsorbers has been developed in
SJTU [10], in which activated carbonmethanol has
been used as the adsorption pair. Shown as Fig. 12, the
system has two adsorbers, each of which has 26 kg car-
bon embedded, the plate n type adsorber makes the
heating and cooling for adsorbers quite quickly. The
system can be operated with a cycle time as short as 20
min. The cooling is output to a fan-coil (actually two
fan-coils of which each has 5 kW cooling power were
used for the experiment) and a cooler is used to cool the
thermal uid of adsorber by cooling water from the
cooling water tower.
Due to the big heat capacity ratio (R=11), the system
COP is low. As an improvement, a novel type adsorber
has been designed, which is a plate nned shell and tube
heat exchanger [10], The designed heat capacity ratio
between adsorber mass plus thermal uid and adsorbent
bed is 2.9 when water is used as the thermal uid; how-
ever, good heat transfer is still possible to reach. Two
new adsorbers were installed in the adsorption air con-
ditioning system, each of them contained 26 kg acti-
vated carbon.
Some tests were performed to the improved adsorp-
tion air conditioning system with novel congured
adsorbers. After various experiments, it was found that
the best cycle time for this adsorption air conditioning
system is between 4050 min with a heat recovery time
for about 2 min. Several typical experiments were done
with the following boundary conditions: room tem-
perature: 24

C, cooling water temperature: 23.5

C and
ow rate of chilled water: 1.157 m
3
/h, thus the perfor-
mances of the system can be evaluated.
The measuring schemes are: (1) Heat source tem-
perature (T
h
): 100

C, evaporation temperature (T
e
):
10

C; cycle time (t): 50 min. (2) T

h
=100

C, T
e
=10

C,
t=40 min. (3) T
h
=100

C, T
e
=6

C, t=40 min. (4)

T
h
=110

C, T
e
=6

C, t=40 min.
Various methods have been used to control the eva-
poration temperature and heat source temperature
properly. The cooling power is evaluated by the aver-
aged temperature dierence of inlet and outlet of chilled
water in the evaporator multiplied by the mass ow rate
of the chilled water and the specic heat of water. The
COP of the system is evaluated as the cooling capacity
divided by the heat input, in which the electric heat
input can be easily measured.
Modications regarding the heat dissipation of the
heating line and boiler to the ambient and the heat leak
to the evaporator were carried out by experiment. Fur-
ther calculation regarding the dierence of heat capacity
of thermal uid between oil and water were also per-
formed (currently water was used as thermal uid;
however, oil is a better uid for the system). The real
system performances with respect to the above four
operation conditions are shown in Table 1. Table 2
shows the predicted performance of the system based
upon experimental results.
Fig. 10. Comparison of COP for various operation procedures
of heat and mass recovery cycle. T
e
=5

C, T
c
=T
a
=30

C,
R=0. 1, basic type cycle; 2, mass recovery cycle; 3, sensible
heat recovery; 4, sensible+adsorption heat recovery; 5, sensible
heat and mass recovery; 6, sensible+adsorption heat and mass
recovery.
Fig. 10. Comparison of COP for various operation procedures of
heat and mass recovery cycle. T
e
=5

C,T
c
=T
a
=30

C, R=0. 1,
basic type cycle; 2, mass recovery cycle;3, sensible heat recovery;
4, sensible+adsorption heat recovery; 5, sensible heat and mass
recovery; 6, sensible+adsorption heat and mass recovery.
Fig. 11. COP of heat and mass recovery cycles with respect to
two dierent heat capacity ratios at T
e
=5

C, T
c
=T
a
=30

C.
Fig. 11. COP of heat and mass recovery cycles with respect to
two dierent heat capacity ratios at T
e
=5

C,T
c
=T
a
=30

C.
R.Z. Wang / International Journal of Refrigeration &(&&&&) && 7
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
The above operation is not completely ideal: heat
recovery is eective only for sensible heat, mass recov-
ery has been adopted and was found eective. As is
shown in Fig. 13, the recorded data for heat recovery
adsorption air conditioning system with and without
mass recovery are quite dierent in Clapeyron diagram,
the cycled mass is extended for the mass recovery cycle.
The further improvement is the heat recovery of
adsorption heat. By the way it was observed that ther-
mal conductivity in the bed is very critical, the heat of
adsorption was not transferred eectively, the adsorp-
tion temperature was about 45

C, which is still very

high. If the eective heat transfer coecient in the
Fig. 12. Schematic of the whole adsorption air conditioning system.
Fig. 12. Schematic of the whole adsorption air conditioning system.
Table 1
Concluded performances of the adsorption air conditioning
with respect to four operation conditions
Tableau 1
Concluded performances of the adsorption air conditioning with
respect to four operation conditions
T
h
(

C)
t
(min)
T
d
(

C)
T
a
(

C)
T
c
(

C)
T
e
(

C)
Q
f
a
(kW)
SCP
1
b
(W/kg)
SCP
2
c
(W/kg)
COP No.
100 50 98.7 46.9 24.0 9.6 3.80 159 146 0.4 1
100 40 96.8 45.6 29.6 9.9 3.93 168 151 0.37 2
100 40 97.8 44.2 26.8 6.1 3.46 148 133 0.34 3
110 40 106 45.7 28.7 6.0 3.70 159 143 0.32 4
a
Q
f
, averaged refrigeration power.
b
SCP
1
, adsorption pair specic cooling power.
c
SCP
2
, system specic cooling power.
Table 2
Several measured performances and its prediction after improv-
ing insulation and substituting heat medium (water to oil)
Tableau 2
Several measured performances and its prediction after improving
insulation and substituting heat medium (water to oil)
Operation condition
No. 1 No. 2 No. 3 No. 4
Performances
SCP
water
measured 159 168 148 159
COP
water
measured 0.40 0.37 0.34 0.32
SCP
water
, real 166 171 151 161
COP
water
, real 0.43 0.40 0.37 0.34
SCP
oil,
predicted 166 171 151 161
COP
oil,
predicted 0.50 0.47 0.44 0.39
Fig. 13. Clapeyron diagram with heat recovery process (- - - -)
and with heat recovery process and mass recovery process ( ).
Fig. 13. Clapeyron diagram with heat recovery process (- - - -)
and with heat recovery process and mass recovery process ( ).
8 R.Z. Wang / International Journal of Refrigeration &(&&&&) &&
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
adsorber was increased, the system performances will be
improved signicantly.
5. Comparisons and discussion
The above experimental results are based upon mass
recovery operation followed after sensible heat recovery.
Theoretical calculations were performed and it was
found that a good agreement between theory and
experiment has been reached, which is shown in Table 3.
Based upon the experimental work and also the the-
oretical results shown in Figs. 9 and 10, it is possible to
predict the maximum COP of the prototype adsorption
air-conditioning system. It was found that with a heat
source temperature of 100

C, it is possible to achieve
COP above 0.5 in real adsorption air conditioning sys-
tem operated at an evaporation temperature about 5

C.
The above work shows clearly that the operation
procedures will have a strong inuence on the COP of
an adsorption air conditioning system, good control is
thereby important to yield better performances.
Mass recovery is very simple to operate, but really
very eective. For operating conditions such as high
condensing temperatures, low evaporation tempera-
tures, or low generation temperatures, mass recovery
operation is strongly recommended. Also, the thermal
capacity ratio between adsorber material and thermal
uid and adsorbent, R, will obviously inuence the
COP. The low R value for a real system will contribute
to increase COP signicantly if mass recovery is used.
Acknowledgements
This work was supported by the state Key Funda-
mentakl Research Program under contract No.
G200026309, the Training Program for the Talents of
State Ministry of Education, and Shanghai Commission
of Science and Technology. The support from Carrier
Corporation and UTRC is especially appreciated. The
author thanks greatly Mr. C.M. Lin, for some calcula-
tions, Ms. J.Y. Wu and Mr. Y.X. Xu for some experi-
ments and discussions.
References
[1] Shelton SV, Wepfer WJ. Solid-vapor air conditioning
technology. IEA Air conditioning Conference 1990;
Tokyo: 52535.
[2] Critoph RE. A forced convection regenerative cycle using
the carbon-ammonia pair. Proc. of the Symposium: Solid
Sorption Refrigeration 1992: 97102.
[3] Douss N, Meunier FE, Sun LM. Predicative model and
experimental result for a two-adsorption air conditioning.
Ind Eng Chem Res 1988;27(2):3106.
[4] Wang RZ. Study on a four-beds cascade adsorption
refrigeration cycle capable of COP over 1.1. Proc. of the
20th International Congress of Refrigeration, 1924 Sep-
tember 1999, Syndey.
[5] Yang LM, Wang RZ, Wu JY. Study on the thermal wave
cycle in the adsorption refrigeration. Journal of Engineer-
ing Thermophysics (in Chinese) 1997;18(6):6736.
[6] Pons M, Poyelle F. Adsorptive machines with advanced
cycles for air conditioninging or cooling applications.
International Journal of Refrigeration 1999;22(1):2737.
[7] Teng Y, Wang RZ, Wu JY. Study of the fundamentals of
adsorption systems. Applied Thermal Engineering
1997;17(4):32738.
[8] Wang RZ, Wang QB. Adsorption Mechanism and
improvements of adsorption equation for adsorption
refrigeration pairs. International Journal of Energy
Research 1999;23:88798.
[9] Wang RZ, Jia JP, Zhu YH, Teng Y, Wu JY, Cheng J,
Wang QB. Study on a new solid adsorption refrigeration
pair: Active carbon ber-methanol. ASME Journal of
Solar Energy Engineering 1997;119(3):2148.
Table 3
Comparisons between theory and experiments for mass and sensible heat recovery and its prediction for the maximum COP in
operation with full heat and mass recovery
Tableau 3
Comparisons between theory and experiments for mass and sensible heat recovery and its prediction for the maximum COP in operation
with full heat and mass recovery
Experiment T
g2
(

C)
T
a
(

C)
T
c
(

C)
T
e
(

C)
COP
(experiment)
COP
(theory)
Predicted maximum
COP with full heat
and mass recovery
R
No. 1 98.7 46.9 24 9.6 0.43 0.46 0.54 2.9
No. 2 96.8 45.6 29.6 9.9 0.4 0.43 0.48
No. 3 97.8 44.2 26.8 6.1 0.37 0.41 0.46
No. 4 105.9 45.7 28.7 6 0.34 0.39 0.44
No. 1 98.7 46.9 24 9.6 0.5 0.51 0.62 1.85
No. 2 96.8 45.6 29.6 9.9 0.47 0.48 0.55
No. 3 97.8 44.2 26.8 6.1 0.44 0.47 0.53
No. 4 105.9 45.7 28.7 6 0.39 0.45 0.51
R.Z. Wang / International Journal of Refrigeration &(&&&&) && 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112
U
n
c
o
r
r
e
c
t
e
d
P
r
o
o
f
[10] Wang RZ, Xu YX, Wu JY, Wang W. Experiments on
heat regenerative adsorption refrigerator and air con-
ditioning. International Journal of Energy Research
1998;22:93541.
[11] Wang RZ, Wu JY, Xu YX, Wang W. Performance
researches and improvements on heat regenerative
adsorption refrigerator and air conditioning. Energy
Conversion and Management 2000: in press.
10 R.Z. Wang / International Journal of Refrigeration &(&&&&) &&
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
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
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
110
111
112