Simulation and analysis of a novel liquid desiccant air-conditioning system

Abstract

In this work a novel energy efficient air-conditioning system utilizing lithium chloride (LiCl) solution as liquid desiccant has been proposed and simulated. The simulation of this system is mainly formulated with two packed columns, one for regenerating the weak desiccant and the other for the dehumidification of ambient air. The air is first dehumidified in the dehumidifier and then sensibly cooled in the indirect and direct evaporative coolers. First and second laws of thermodynamics have been used to analyze the effect of five key variables on the performance of the system. High efficiency could be achieved if proper values of these variables are selected.

Full text

Simulation and analysis of a novel liquid desiccant air-conditioning system
Min Tu
a
, Cheng-Qin Ren
a,*
, Long-Ai Zhang
b
, Jian-Wei Shao
c
a
College of Mechanical and Automotive Engineering, Hunan University, Lushan road, Changsha, Hunan 410082, China
b
Gree Electric Appliances, Inc. of Zhuhai, 519070 Guangdong, China
c
China Machinery International Engineering Design and Research Institute, China
article info
Article history:
Received 13 May 2007
Accepted 15 December 2008
Available online 24 December 2008
Keywords:
Liquid desiccant system
Air-conditioning
Energy efficiency
First law
Second law
abstract
In this work a novel energy efficient air-conditioning system utilizing lithium chloride (LiCl) solution
as liquid desiccant has been proposed and simulated. The simulation of this system is mainly
formulated with two packed columns, one for regenerating the weak desiccant and the other for
the dehumidification of ambient air. The air is first dehumidified in the dehumidifier and then sensi-
bly cooled in the indirect and direct evaporative coolers. First and second laws of thermodynamics
have been used to analyze the effect of five key variables on the performance of the system. High
efficiency could be achieved if proper values of these variables are selected.
Ó2008 Elsevier Ltd. All rights reserved.
1. Introduction
In hot and humidity climates, one of the most insistent de-
mands is air-conditioning (refrigeration and dehumidification) in
occupied spaces, and a large share of electric energy has been con-
sumed by the conventional vapor-compression air-conditioning
systems. In these areas, the air humidity ratio is high, which is
not desirable for human comfort. Process air is passed over a
low-temperature cooling coil to remove some of its moisture. As
a result, the conventional vapor-compression air-conditioning sys-
tems must normally operate at a temperature colder than the sup-
ply air dew-point temperature.
A promising alternative energy saving air-conditioning method
is introduced. Different from the conventional vapor-compression
air-cooling system, the system deals with the sensible load and
the latent load, respectively. The sensible cooling of the air is
accomplished at temperatures above the wet bulb. The latent part
of the cooling load can be treated by using desiccants. Desiccants
have a property of absorbing water vapor, and they can be solid
or liquid. Factor and Grossman [1] listed some other advantages
of the liquid desiccant air-conditioning system as follow: ease of
manipulation, low pressure drop in the contactors with the air,
possibility of filtering to remove dirt taken in from the air, possibil-
ity of heat exchange between spent and regenerated desiccant
streams and more, also requiring lower temperature for regenera-
tion. Thus, the following analysis is focused on the liquid desiccant
systems.
The concept of using liquid desiccant system to dehumidify the
air goes back to the year of 1955 when Löf designed an open-cycle
air-conditioning system using triethylene glycol as liquid desiccant
solution. Since then, the technique has been further developed and
studied by more and more investigators [2–12]. The liquid desic-
cant air-conditioning systems mainly consist of two types: the
pure liquid desiccant system and the hybrid.
Many research works can be found on pure liquid desiccant sys-
tem. Scalabrin and Scaltriti [2] presented and simulated an open
process of summer air conditioning by dehumidification–humidifi-
cation. An internally cooled dehumidifier and an internally heated
regenerator composed of the main components of the proposed
system. This system did not utilize the cooling capacity of exhaust
air from air-conditioning space. Ameel et al. [3] investigated an
open-cycle-absorption-refrigeration (OCAR) system where the
absorber was internally cooled. Sanjeev et al. [4] did experimental
studies on a liquid desiccant system, in which a falling film tubular
absorber and a falling film plate regenerator formed the heart of
that system. But it is regrettable that the system’s performance
and characteristics had not been analyzed in detail. Hellmann
and Grossman [5] simulated and analyzed an open-cycle dehumid-
ifier-evaporator-regenerator (DER) absorption chiller for low-grade
heat energy utilization. Pohl et al. [6] compared a semi-open mode
with the open mode by Grossman [5]. The two modes consist of
only three major components: an internally cooled dehumidifier,
an internally heated regenerator and an evaporative cooler. As
extended researches, Gommed et al. [7], Gommed and Grossman
[8,9] did experimental investigations on the performance of the
open absorption system, in which the dehumidifier and regenera-
tor operated in an adiabatic mode. In all the systems proposed by
1359-4311/$ - see front matter Ó2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2008.12.006
*Corresponding author. Tel.: +86 13874886953; fax: +86 731 8711911.
E-mail address: renchengqin@163.com (C.-Q. Ren).
Applied Thermal Engineering 29 (2009) 2417–2425
Contents lists available at ScienceDirect
Applied Thermal Engineering
journal homepage: www.elsevier.com/locate/apthermeng

Hellmann and Grossman [5], Pohl et al. [6], Gommed et al. [7],
Gommed and Grossman [8,9], chilled water was produced to cool
the process air below its dew-point temperature for removing
some of its moisture, so the process air must be reheated before
entering the residential space.
Recently, research works on the hybrid liquid desiccant air-con-
ditioning systems can also be found on published literatures. Adnan
et al. [10,11] proposed and analyzed the performances of a hybrid
system, which composed of a liquid desiccant air-dehumidification
system and a conventional vapor-compression air-cooling system.
But the configuration of that system was so complex that the initial
cost was very large. Ahmed et al. [12] simulated a hybrid open-cycle
vapor absorption and liquid desiccant system using LiBr for the pro-
cess of absorption and dehumidification. The cooling water was pro-
duced in the absorber/evaporator component and thus the exhaust
air from the residential space can not be utilized adequately.
From the review of the previous works we can get that: some
systems did not utilize the cooling capacity of exhaust air from
air-conditioning space; some ones produced chilled water to cool
the process air, so the reheat of the process air might also be
needed; and the others’ configurations might be too complex.
In the present work, a modified liquid desiccant air-conditioning
system is introduced and illustrated in Fig. 1. This system is mainly
composed of a dehumidifier, a regenerator, an indirect evaporative
cooler and a direct evaporative cooler. It in effect separates the sen-
sible and the latent cooling loads. The process air is first dehumidi-
fied in the dehumidifier to overcome the latent load, and then
exhaust air from air-conditioning space is utilized as the secondary
air to cool the process air in the indirect evaporative cooler to over-
come the sensible load. The weak solution leaving the dehumidifier
is re-concentrated in the regenerator, and then re-circulated back to
the dehumidifier. Thus, a continuous supply of strong desiccant
solution to dehumidify the process air can be assured. The present
work has two more purposes: developing a modular simulation pro-
gram for the following analysis and accomplishing a detailed analy-
sis of the effects of several key parameters on the performance of this
system based on the first and second laws of thermodynamics.
2. System description
The proposed system in this paper is shown in Fig. 1. This liquid
desiccant system contains three loops: the process air loop, the
liquid desiccant solution loop and the air loop in regeneration.
The process air loop is composed of the dehumidifier, the indi-
rect evaporative cooler (IEC) and the direct evaporative cooler
(DEC). In this loop, process air from the atmosphere (state point
7) is first dehumidified by the dehumidifier to state point 8. Then
it is cooled to state point 9 without any moisture content variation
in the IEC, where the exhaust air from the air-conditioning space is
utilized as the secondary air. Finally, the process air is cooled in the
DEC adiabatically to state point 10. Because the cooling energy of
exhaust air is already recovered in IEC to the process air, air recir-
culation is not necessary.
Some researchers suggested that using outlet secondary air to
cool the inlet secondary air of the indirect evaporative cooler with
an air-to-air heat exchanger can improve the performance of the
Nomenclature
Cspecific heat, [kJ/(kg °C)]
C
*
air-to-solution heat capacity ratio
C
pa
specific heat of air, [kJ/(kg °C)]
C
pv
specific heat of water vapor, [kJ/(kg °C)]
C
pw
specific heat of water, [kJ/(kg °C)]
COP coefficient of performance
D
e
mass transfer coefficient, D
e
=H/C
pa
[kg/(m
2
s)]
dhumidity ratio, [kg/kg.d.a]
Ex exergy of energy transferred, [kJ]
Ex
Qcd
exergy of cooling capacity, [kJ]
Ex
Qgen
exergy of heat energy for regeneration, [kJ]
especific enthalpy, [kJ/kg d a]
e
m
(t) enthalpy of water vapor, [kJ/kg]
ex exergy of moist air, [kJ/kg]
Hheat transfer coefficient, H=54V
0.7
[kW/(m
2
°C)]
h
s
heat of absorption, [kJ/kg]
~
hsnormalized heat of absorption, ~
hs¼Mwhs=MaCpa½C
Le Lewis factor for the overall heat and mass transfer pro-
cesses
Mmolecular weight, [kg/kmol]
mmass flow rate, [kg/s]
m
a
dry air mass flow rate, [kg/s]
m
L
mass flow rate of solution, [kg/s]
NTU number of heat transfer units
ppressure, [kPa]
Qsensible heat transferred, [kJ]
R
a
gas constant for dry air, [kJ/(kg °C)]
R
va
water vapor to dry air specific heat capacity ratio,
R
va
=M
w
C
pv
/M
a
C
pa
R
dw
dehydrated-desiccant-to-water molecular weight ratio,
R
dw
=M
d
/M
w
R
m
air-to-dehydrated-desiccant mass flow rate ratio,
R
m
=m
a
M
d
/(m
d
M
a
)
TKelvin temperature, [K] (t+ 273.15 K)
tCelsius temperature of fluid, [°C]
Vvelocity, [m/s]
ylength of the primary air flow direction, [m]
zlength of the secondary air flow direction, [m]
Greek letters
D
change of or difference between parameters
eheat transfer efficiency of the heat exchanger
g
second law efficiency (the exergy efficiency)
nmass fraction of desiccant in solution, (wt% salt)
w
wettability factor of the plate
Subscripts
0 ambient air state or dead state
1, 2, 3, 4... state points
a air
c cold
cd cooling capacity
d desiccant
e at equilibrium condition
eL at equilibrium condition with solution
gen regeneration
HE heat exchanger
h hot
i inlet
L solution or liquid phase
LD liquid desiccant system
max maximum
min minimum
o outlet
p primary air
s secondary air
sat saturate state
w water or water film
2418 M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425

system [6]. But it is not appropriate for the proposed system in this
article. By the analysis of Ref. [13], the changes of the secondary air
inlet temperature have little effect on the primary air outlet tem-
perature of IEC. Simultaneously, for decreasing the initial cost, we
did not prefer to add an air-to-air heat exchanger before the IEC.
The solution loop consists of a water cooler, the dehumidifier, a
solution-to-solution heat exchanger HE2, a heater and the regener-
ator. The weak solution flows out of the dehumidifier (state point
1) is first pre-heated by HE2to state point 2, latter heated in the
heater, finally re-concentrated in the regenerator. The re-concen-
trated solution (state point 4) is pre-cooled to state point 5 by
HE2then cooled in the water cooler to state point 6, and sprayed
into the dehumidifier at last. The cooling water in the water cooler
is supplied by a cooling tower, which can decrease the strong solu-
tion temperature only 3 °C higher than the wet bulb temperature
of the ambient air, by adjusting the flux of the recirculation water.
The air loop in regeneration is composed of the air-to-air heat
exchanger HE1and the regenerator. The air (state point 13, the
same as state point 7) is first pre-heated in HE1to state point 14,
and then fed into the regenerator, absorbing water vapor from
weak solution. The air flowing out of the regenerator (state point
15) transfers the heat to the air of state 13 in HE1. And then it is
discharged to the outside.
Is there any need to add an air-to-air heat exchanger HE1before
the regenerator? Ertas et al. [14] thought that: this reveals that an
air pre-heater or an air-to-air heat exchanger placed between the
entering and leaving air streams may decrease the COP value of a
desiccant system. But from the reference [15], one can get that:
the rate of water evaporation increases, while the temperature of
air supplied to the regenerator increases, deservedly the regenera-
tion efficiency increases. Consequently, an air-to-air heat exchan-
ger is needed to improve the regeneration efficiency and reduce
the thermal losses.
Simulation results of several important air and solution state
points are shown on the psychrometric chart of Fig. 2.
Direct
Evaporative
Cooler
Indirect
Evaporative
Cooler Dehumidifier
Cooler
Regenerator
Heater
HE2
Cooling
Wa t e r
HE1
3
41
2
6
7
8
5
1
9
10
12
11
14
13
(
7
)
19
16
15
20
17
18
Heating
Water
Air loop in
regeneration
Air-conditioning
loop
Liquid desiccant
solution loop
Fig. 1. Description for the proposed liquid desiccant air-conditioning system.
Fig. 2. Simulation results for several important state points of air and solution.
M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425 2419

3. Mathematical modeling
The aqueous lithium chloride desiccant air-conditioning system
proposed in this paper is composed of several sub-units. In the
beginning, each subsystem is modeled and then a computer code
is prepared, debugged and tested to evaluate the performance of
the whole liquid desiccant air-conditioning system. For simplifica-
tion of simulation, some assumptions are adopted as following:
(1) the system operates in steady state;
(2) the power consumption of pumps and fans is neglected.
3.1. Packed beds
The packed beds’ finite difference model developed by Ren et al.
[16] is adopted in this work. Packing material has been filled into
packed columns. The solution is sprinkled over the packing mate-
rial at the top of the pack bed tower, where it comes in direct con-
tact with a counter-current flow air. The governing equations for
the coupled heat and mass transfer processes can be obtained by
following the principles of energy and mass conservation. Then,
those equations are rearranged in terms of some dimensionless
and dimensional parameter groups, obtaining equations (1)–(5):
dd
a
¼ðd
a
d
eL
Þ1
Le dNTU ð1Þ
dn¼1:608ðd
a
d
eL
ÞR
m
n
2
Le½ð2R
2
dw
2R
dw
Þn2R
2
dw
þR
dw
dNTU ð2Þ
dT
a
¼ðT
a
T
L
ÞLe 1:608R
v
a
ðd
a
d
eL
Þ
Leð1þ1:608R
v
a
d
a
ÞdNTU ð3Þ
dT
L
¼C

ðT
a
T
L
Þþ1:608
~
h
s
Le ðd
a
d
eL
Þ
"#
dNTU ð4Þ
dm
L
¼m
a
dd
a
ð5Þ
The details of the above equations for the coupled heat and mass
transfer processes were formulated in reference [16]. For simulating
of the packed beds, the number of heat transfer units and the Lewis
factor for the overall heat and mass transfer processes are equal to
3.0 and 1, respectively.
3.2. Indirect evaporative cooler
Several models of the indirect evaporative cooler have been
proposed in some previous works. The model used in this paper
is based on the model proposed in reference [17] after some
corrections in presentations have been made in reference [13].A
summary of this model is presented in the following equations
(6)–(10):
dt
p
¼H
p
ðt
p
t
w
Þ
C
pa
þ1:85d
p
dz
m
p
dyð6Þ
dm
w
¼D
e
ðd
sat
d
s
Þwdydzð7Þ
dd
s
¼dm
w
m
s
ð8Þ
dt
s
¼e
v
ðt
w
Þe
v
ðt
s
Þ
C
pa
þ1:85d
s
dd
s
þH
s
ðt
w
t
s
Þ
C
pa
þ1:85d
s
dz
m
s
dyð9Þ
dt
w
¼1
m
w
C
pw
m
p
ðC
pa
þ1:85d
p
Þdt
p
m
s
ðC
pa
þ1:85d
s
Þdt
s
þð2501 þ1:85t
s
C
pw
t
w
Þdm
w

ð10Þ
The primary air of this IEC is the air out of the dehumidifier, the
secondary air is the exhaust air of air-conditioning space, and the recy-
cled water is sprayed over the secondary air side. The primary air stream
can be cooled without increasing water content. The interaction
between the air streams and the water film is very complicated, and
detailed descriptions of this mathematic model are expressed in refer-
ence [17]. In this paper, the indirect evaporative cooler is divided into
250 U. In each unit, the length of the primary air flow direction yis equal
to 1 m, so does the length of the secondary air flow direction z.Andthe
value of the wettability factor of the plate
w
is 0.8.
3.3. Direct evaporative cooler
A direct evaporative cooler (DEC) is used to modulate the
humidity ratio of the process air. It is an approximately isoenthalpy
process in the DEC. When the heat/moisture ratio in the air-condi-
tioning space is defined as constant, the supply air state point is the
intersection of isenthalpic and constant heat/moisture ratio line. In
this paper, the heat/moisture ratio is equal to 4000.
3.4. Heat exchangers
Two types of heat exchangers have been used: air-to-air heat
exchanger and solution-to-solution heat exchanger. These heat
exchangers’ flow direction is counter-current, then e-NTU method
can be adopted in calculating the outlet temperatures.
The number of heat transfer units (NTU) is first defined. Then
comparing the values of m
h
C
h
and m
c
C
c
, the larger one is (mC)
max
,
the other is (mC)
min
. From reference [18], the follow equation can
be utilized:
e
¼
1exp ðNTUÞ1
ðmCÞ
min
ðmCÞ
max
hino
1
ðmCÞ
min
ðmCÞ
max
exp ðNTUÞ1
ðmCÞ
min
ðmCÞ
max
hino ð11Þ
where, eis the heat transfer effectiveness of the heat exchanger.
The quantity of transferred heat:
Q
HE
¼
e
ðmCÞ
min
ðt
hi
t
ci
Þð12Þ
The outlet temperature of hot fluid:
t
ho
¼t
hi
Q
HE
m
h
C
h
ð13Þ
The outlet temperature of cold fluid:
t
co
¼t
ci
þQ
HE
m
c
C
c
ð14Þ
In the simulation, the number of heat transfer units of the air-to-air
heat exchanger and the solution-to-solution heat exchanger are de-
fined as 3.0.
4. Analyses
4.1. First law analysis
The first law efficiency of this thermodynamic cycle measures
the fraction of the input energy that is converted to useful energy
output. It is usually defined as COP (Coefficient of Performance).
For this liquid desiccant air-conditioning system, the COP can be
given by:
COP
LD
¼Q
cd
Q
gen
ð15Þ
where, Q
cd
is the cooling capacity which is produced by this liquid
desiccant cooling system; Q
gen
is the heat energy which is required
to regenerate the weak solution, and it is absorbed by weak solution
in the heater.
Q
cd
, the cooling capacity, can be shown to be:
Q
cd
¼ðe
7
e
10
Þm
a7
ð16Þ
2420 M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425

Q
gen
, the heat energy for regeneration, can be written as:
Q
gen
¼ðC
L3
T
3
C
L2
T
2
Þm
L2
ð17Þ
Because of the specific heat C
L
changes very small with the solution
temperature increases from T
2
to T
3
, equation (17) can be rear-
ranged as:
Q
gen
¼C
L2
m
L2
ðT
3
T
2
Þð18Þ
4.2. Second law analysis
Usually, the first law of thermodynamics is used to investigate
the system’s performance. However, the first law is only concerned
with the quantity of energies in conversation without considering
their quality differences. For instance, compared with the same
amount of heat energy, the electrical energy has different capacity
for work. Therefore, a new physical parameter ‘‘exergy” is widely
used to assess both the quantity and quality of the energies. Hasan
et al. [19] used the first and second laws of thermodynamics to
analyze a novel thermodynamic cycle proposed by Goswami in
1995. Ahmed et al. [20] used the exergy analysis to estimate irre-
versible losses of a liquid-desiccant-based, hybrid air-conditioning
system. Thus, a second law efficiency
g
based on exergy balance is
adopted in this article to depict the performances of the proposed
thermodynamic system.
In some literature the second law efficiency
g
, is defined as the
ratio of the exergy output to the exergy input, it is shown to be:
g
¼Ex
output
D
Ex
input
ð19Þ
where,
D
Ex
input
is the change in exergy which is obtained by the
weak solution from heater, it is equal to Ex
Qgen
. The output exergy
Ex
output
is the exergy of cooling capacity; it is equal to Ex
Qcd
in this
proposed system. So the second law efficiency of liquid desiccant
air-conditioning system
g
LD
is given by:
g
LD
¼Ex
Qcd
Ex
Qgen
ð20Þ
From Ref. [21], one can get that when the heat transfers in a tem-
perature variation process, the heat flow exergy is:
Ex ¼Z1T
0
T

dQð21Þ
Because of the weak solution absorbing heat energy Q
gen
in the hea-
ter, its temperature increases from T
2
to T
3
. So, the heat flow exergy
is:
Ex
Qgen
¼Z
T
3
T
2
1T
0
T

dQ
gen
ð22Þ
Using equation (18) and (22), one can arrive at the heat flow exergy
in the heater as:
Ex
Qgen
¼C
L2
m
L2
½T
3
T
2
þT
0
ðln T
2
ln T
3
Þ ð23Þ
For the exergy analysis of moist air, the selection of dead state is
very important. Usually, the atmospheric state (T
0
,p
0
,d
0
) is selected
as the dead state. However, when the ambient air is not saturated, it
still possesses available energy. Therefore, the state (T
0
,p
0
,d
0sat
)is
selected as the dead state. And the moist air’s flow exergy can be
represented as [21]:
ex ¼ðC
pa
þC
p
v
dÞT
0
T
T
0
1ln T
T
0

þð1þ1:608dÞR
a
T
0
ln p
p
0
þR
a
T
0
ð1þ1:608dÞln 1þ1:608d
0sat
1þ1:608dþ1:608dln d
d
0sat

ð24Þ
From equation (24), the exergy of moist air is broken down into
three components: thermal, mechanical and chemical. The pressure
difference of moist air between indoor and outdoor is very small,
and the state point 7 is the ambient air state point, hence the exergy
of supply air (state point 10) is:
ex
10
¼C
pa
þC
p
v
d
10

T
7
T
10
T
7
1ln T
10
T
7

þR
a
T
7
ð1þ1:608d
10
Þln 1þ1:608d
7sat
1þ1:608d
10
þ1:608d
10
ln d
10
d
7sat

ð25Þ
The exergy of ambient air (state point 7) is:
ex
7
¼R
a
T
7
ð1þ1:608d
7
Þln 1þ1:608d
7sat
1þ1:608d
7
þ1:608d
7
ln d
7
d
7sat

ð26Þ
As a result, the exergy of cooling capacity is expressed as follow:
Ex
Qcd
¼m
a7
ðex
10
ex
7
Þð27Þ
5. Results and discussion
The effects of five key parameters on the performance of the
proposed liquid desiccant cooling system have been simulated
and analyzed. These parameters are the inlet solution temperature
in regenerator, the air-to-dehydrated desiccant mass flow rate
ratio m
a
/m
d
in dehumidifier and regenerator, ambient air temper-
ature and relative humidity. Numerical simulations are conducted
within the following ranges of parameters: 55–85 °C for the inlet
solution temperature to the regenerator, 0.303–7.245 [kg/kg] for
m
a
/m
d
in the dehumidifier, 0.658–2.108 [kg/kg] for m
a
/m
d
in the
regenerator, 25–43 °C for the ambient air temperature and 37–
90% for the ambient air relative humidity. As a reference case of
simulation, these parameters’ design points are shown in Table 1.
A sensitivity analysis is performed by varying the parameters of
interest one at a time, while keeping all others fixed at their design
values. Room air conditions are selected as 26 °C of temperature
and 60% of relative humidity, which are typical in summer air con-
ditioning design.
5.1. Effect of inlet solution temperature to the regenerator
The effects of changing the inlet solution temperature to the
regenerator on the performance of the liquid desiccant air-condi-
tioning system are shown in Fig. 3.
As shown in Fig. 3(a), the humidity ratio and temperature of the
supply air decrease as the inlet solution temperature in the regen-
erator increases. The explanation for this is as follow: a higher inlet
solution temperature to the regenerator gets a higher outlet solu-
tion concentration, thus a lower equilibrium humidity ratio of
solution results. It causes an enhancement in absorption perfor-
mance and leads to lower water content in the supply air. After
being cooled in an approximately isoenthalpy process by DEC,
the temperature of supply air (state point 10) will be further
lowered.
Table 1
System parameters for reference case.
Parameter Value
Inlet solution temperature to the regenerator (°C) 75
Ambient air temperature (°C) 35
Ambient air relative humidity 60%
Mass flux of the dry air in packed beds (kg/s) 1.0
Mass flux of the dehydrated solution in packed beds (kg/s) 0.76
M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425 2421

Fig. 3(b) describes the cooling capacity and COP of the liquid
desiccant air-conditioning system as the functions of the inlet solu-
tion temperature in the regenerator. As shown in Fig. 3(a), the
humidity ratio and temperature of supply air (state point 10)
decrease at higher inlet solution temperature in the regenerator,
so the enthalpy difference between states points 7 and 10 increases.
Then, from equation (16), we can find that the cooling capacity Q
cd
increases too. Although the humidity ratio of the supply air
decreases all the while, but for the mass flow rate of process air keeps
constant, the water absorbed by strong solution in the dehumidifier
is still small. That means the large part of heat energy absorbed by
weak solution in the heater is used to heat the solution and air in
the regenerator, and the small part is really used to re-concentrate
the weak solution. So, when the inlet solution temperature to the
regenerator increases, there will be more and more heat energy used
to heat the solution and air, and the COP decreases.
Fig. 3(c) shows the exergy of cooling capacity and the exergy
efficiency of this liquid desiccant air-conditioning system as func-
tions of the inlet solution temperature in regenerator. Obviously,
the higher the inlet solution temperature in regenerator means
the higher the Ex
Qcd
. The exergy efficiency
g
LD
decreases signifi-
cantly at the lower ranges of the inlet solution temperature to
the regenerator, and then slowly down when the temperature is
more than 75 °C.
For the regenerator of this liquid desiccant system, a higher in-
let solution temperature means a higher concentration of outlet
solution. But if the inlet solution temperature to the regenerator
exceeds 85 °C, the solution in the water cooler will not stay safely
away from the crystallization limit. According to the results of pre-
vious simulation, 80 °C is a safe solution temperature for regener-
ation, and it is easy to get by plate-type solar energy collectors.
5.2. Effect of the air-to-dehydrated desiccant mass flow rate ratio in
the dehumidifier
The analysis is performed by changing the mass flux of air in
dehumidifier but keeping the other variables constant. The effects
of changing the air-to-dehydrated desiccant mass flow rate ratio
m
a
/m
d
in the dehumidifier are shown in Fig. 4.
Fig. 4(a) describes that the supply air humidity ratio and tem-
perature increase with the air-to-dehydrated desiccant mass flow
rate ratio. Higher value of m
a
/m
d
means lower heat and mass
capacities of solution, this will lead to a smaller average heat and
mass transfer potential for moist air, thus the higher outlet tem-
perature and humidity ratio of process air result.
Fig. 4(b) shows the cooling capacity and COP of the liquid des-
iccant air-conditioning system as functions of the air-to-dehy-
drated desiccant mass flow rate ratio m
a
/m
d
. From Fig. 4(a), we
can get that the supply air humidity ratio and temperature increase
with the increase of m
a
/m
d
values, then the enthalpy difference be-
tween supply air and ambient air decreases. But because of the air
mass flux m
a7
increases quickly, the cooling capacity Q
cd
mount up
all the while. And at the same time, Q
cd
increases more rapidly than
Q
gen
, therefore the COP goes up all along.
Fig. 4(c) describes the effect of changing m
a
/m
d
in dehumidifier
on the exergy of cooling capacity and exergy efficiency of this air-
conditioning system. From the observation of this figure, we can
get that Ex
Qcd
increases rapidly at the lower ranges of air-to-dehy-
drated desiccant mass flow rate ratio but increases more slowly
when the value of m
a
/m
d
is more than 5.27 [kg/kg]. The exergy effi-
ciency of this liquid desiccant system
g
LD
has the same tendency
with Ex
Qcd
.
It should be noticed that the air-to-dehydrated desiccant mass
flow rate ratio should be restricted to make the supply air humidity
or temperature below a certain value. For example, if the supply air
temperature is needed below 20 °C, the air-to-dehydrated desic-
cant mass flow rate ratio m
a
/m
d
in the dehumidifier must be less
than 2.0 [kg/kg].
5.3. Effect of the air-to-dehydrated desiccant mass flow rate ratio in
the regenerator
The analysis is performed by changing the mass flux of air in
regenerator but keeping the other variables constant. The effects
of changing the air-to-dehydrated desiccant mass flow rate ratio
m
a
/m
d
in regenerator are shown in Fig. 5.
Fig. 5(a) describes the supply air humidity ratio and tempera-
ture as functions of the m
a
/m
d
in the regenerator. When the value
of m
a
/m
d
increases from 0.658 to 2.108 [kg/kg], the more the air
blown into regenerator the more the water evaporating from weak
solution, and the concentration of solution out of the regenerator
increases. Thus the humidity ratio of supply air decreases. How-
Inlet solution temperature in regenerator [ºC]
Supply air temperature, t10 [ºC]
Supply air humidity ratio, d10 [kg/kg]
0
5
10
15
20
25
50 60 70 80 90
0
0.002
0.004
0.006
0.008
0.01
0.012
10
t
10
d
Coefficient of performance, COP
Inlet solution temperature in regenerator [ºC]
Cooling capacity, Qcd [kJ]
0
10
20
30
40
50
60
50 60 70 80 90
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
COP
cd
Q
Inlet solution tem
p
erature in re
g
enerator [ºC]
Exergy efficiency,
Exergy of cooling capacity, ExQcd [kJ]
0
0.5
1
1.5
2
2.5
3
50 60 70 80 90
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
LD
η
Qcd
Ex
LD
a
b
c
η
Fig. 3. The effects of the inlet solution temperature to the regenerator on the
performance of the liquid desiccant air-conditioning system.
2422 M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425

ever, for values of m
a
/m
d
being greater than 1.6 [kg/kg], the humid-
ity ratio of supply air will decrease slowly. The reason is that the
weak solution is regenerated almost sufficiently when the value
of m
a
/m
d
is equal to 1.6 [kg/kg]. After being cooled by IEC and
DEC, the temperature of the supply air decreases with the same
tendency as the supply air humidity ratio.
Fig. 5(b) describes the cooling capacity and the system’s coeffi-
cient of performance (COP) as functions of m
a
/m
d
in the regenera-
tor. From Fig. 5(a) and equation (16), when the humidity ratio and
temperature of supply air decrease, the enthalpy difference be-
tween state points 7 and 10 increases, thus the cooling capacity
Q
cd
increases. Owing to the increasing of air mass flux in the regen-
erator, the average mass transfer capacity between air and weak
solution increases. So the needed energy for regeneration Q
gen
de-
creases and the COP increases.
Fig. 5(c) shows that the exergy of cooling capacity Ex
Qcd
and
exergy efficiency
g
LD
increase rapidly when the value of m
a
/m
d
is
below 1.6 [kg/kg]. But above this value both of them increase
slowly.
From the above three figures, one can see that continuously
increasing the mass flux of ambient air to the regenerator has little
effect on increasing the system’s performance, when the weak
solution is regenerated adequately.
5.4. Effect of ambient air temperature
The effects of changing the ambient air temperature on the per-
formance of this liquid desiccant air-conditioning system are
shown in Fig. 6. By changing the ambient air temperature while
keeping the air relative humidity constant, the air absolute humid-
ity is varied.
da
mm / in the dehumidifier, [kg/kg]
Supply air temperature, t
10
[ºC]
Supply air humidity ratio, d
10
[kg/kg]
10
12
14
16
18
20
22
24
26
28
30
02468
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
10
t
10
d
Coefficient of performance, COP
da
mm / in the dehumidifier, [kg/kg]
Cooling capacity, Qcd [kJ]
0
50
100
150
200
250
02468
0
0.2
0.4
0.6
0.8
1
1.2
1.4
COP
cd
Q
da
mm / in the dehumidifier, [kg/kg]
Exergy of cooling capacity, Ex
Qcd
[kJ]
0
1
2
3
4
5
6
7
8
02468
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
LD
η
Qcd
Ex
a
b
c
Exergy efficiency, LD
η
Fig. 4. The effects of the air-to-dehydrated desiccant mass flow rate ratio in the
dehumidifier on the performance of the liquid desiccant air-conditioning system.
da mm / in the regenerator, [kg/kg]
Supply air temperature, t10 [ºC]
Supply air humidity ratio, d10 [kg/kg]
15
16
17
18
19
20
21
22
00.5 11.5 22.5
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
10
t
10
d
Coefficient of performance, COP
da mm / in the regenerator, [kg/kg]
Cooling capacity, Qcd [kJ]
47
48
49
50
51
52
53
54
0 0.5 1 1.5 2 2.5
0.4
0.41
0.42
0.43
0.44
0.45
0.46
0.47
0.48
0.49
0.5
COP
cd
Q
da mm / in the regenerator, [kg/kg]
Exergy of cooling capacity, ExQcd
1.8
1.9
2
2.1
2.2
2.3
2.4
0 0.5 1 1.5 2 2.5
0.1
0.15
0.2
0.25
0.3
0.35
LD
ηLD
η
Qcd
Ex
[kJ]
a
b
c
Exergy efficiency, LD
η
Fig. 5. The effects of the air-to-dehydrated desiccant mass flow rate ratio in the
regenerator on the performance of the liquid desiccant air-conditioning system.
M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425 2423

From Fig. 6(a), one can see that the humidity ratio and temper-
ature of supply air increase with the increase of ambient air tem-
perature values. The explanation for this is as follow: First, when
the ambient air relative humidity keeps constant, higher air tem-
perature means higher absolute humidity of air, so the water
content of the air blown into the regenerator increases, thus the
concentration of strong solution out of the regenerator decreases.
Second, because of the ambient air wet bulb temperature
increases, the inlet solution temperature to the dehumidifier
increases. As a result, the dehumidifier outlet air humidity ratio
and temperature increase.
Fig. 6(b) shows the cooling capacity and COP as functions of
ambient air temperature. Although the supply air temperature
and humidity ratio increase, the enthalpy difference between the
supply air and the ambient air increases. So, by equation (16),we
can get that the cooling capacity Q
cd
will still increases. Simulta-
neously, as the energy required for regeneration Q
gen
decreases,
the COP increases.
Fig. 6(c) describes the exergy of cooling capacity Ex
Qcd
and exer-
gy efficiency
g
LD
as functions of ambient air temperature. The exer-
gy of cooling capacity increases nearly straightly. The exergy
efficiency increases very slowly at first but quickly at higher ranges
of ambient air temperatures.
Generally, from the above three figures we can get that this li-
quid desiccant system is suit for operating at wide ranges of ambi-
ent air temperature. Even though in the high temperature areas,
the COP and exergy efficiency of this system have the larger values.
5.5. Effect of ambient air relative humidity
The effects of changing the ambient air relative humidity on the
performance of the liquid desiccant system are shown in Fig. 7.By
Ambient air temperature, [ºC]
Supply air temperature, t10 [ºC]
Supply air humidity ratio, d10 [kg/kg]
0
5
10
15
20
25
30
20 25 30 35 40 45 50
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
10
t
10
d
Coefficient of performance, COP
Ambient air temperature, [ºC]
Cooling capacity, Qcd [kJ]
0
10
20
30
40
50
60
70
80
90
20 25 30 35 40 45 50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
COP
cd
Q
Ambient air tem
p
erature, [ºC]
Exergy efficiency,
Exergy of cooling capacity, ExQcd [kJ]
0
0.5
1
1.5
2
2.5
3
3.5
4
20 25 30 35 40 45 50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Qcd
Ex
LD
η
η LD
a
b
c
Fig. 6. The effects of the ambient air temperature on the performance of the liquid
desiccant air-conditioning system.
Ambient air relative humidity
Supply air temperature, t10 [ºC]
Supply air humidity ratio, d10 [kg/kg]
0
5
10
15
20
25
30% 50% 70% 90%
0
0.002
0.004
0.006
0.008
0.01
0.012
10
t10
d
Ambient air relative humidity
Coefficient of performance, COP
Cooling capacity, Qcd
0
10
20
30
40
50
60
70
80
30% 50% 70% 90%
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
COP
cd
Q
Ambient air relative humidit
y
Exergy of cooling capacity, ExQcd [kJ]
0
0.5
1
1.5
2
2.5
30% 50% 70% 90%
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
LD
η
Qcd
Ex
[kJ]
a
b
c
Exergy efficiency, LD
η
Fig. 7. The effects of the ambient air relative humidity on the performance of the
liquid desiccant air-conditioning system.
2424 M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425

maintaining the ambient air temperature but changing the relative
humidity, the air absolute humidity is varied.
Fig. 7(a) describes the supply air humidity ratio and tempera-
ture as functions of the ambient air relative humidity. A higher
ambient air absolute humidity results in a lower outlet solution
concentration from the regenerator. Because of the lower inlet
solution concentration, the higher inlet air absolute humidity and
solution temperature to the dehumidifier, the process air out of
dehumidifier has the higher humidity and temperature. And then
after being cooled orderly in IEC and DEC, the supply air also has
the higher humidity and temperature.
From Fig. 7(b), one can see that the cooling capacity and COP are
nearly linear functions of ambient air relative humidity. They in-
crease as the ambient air relative humidity increases.
Fig. 7(c) shows the exergy of cooling capacity and exergy effi-
ciency initially increase at the lower ranges of ambient air relative
humidity but decrease significantly at the higher ambient air rela-
tive humidity. The values of Ex
Qcd
and
g
LD
reach the maximum at
ambient air relative humidity of 55%.
The liquid desiccant system is suit for operating at wide ranges
of relative humidity conditions. The supply air of this system can
meet the comfortable requirement of residential space, even at
the high relative humidity areas. But at the same time, if the ambi-
ent air relative humidity is small, the process air is not needed to
be dehumidified in the dehumidifier; it will only need to be cooled
in IEC and DEC.
6. Conclusions
Usually the conventional vapor-compression air-cooling systems
use the electric power as the driving force, but the liquid desiccant
air-conditioning system makes it possible to be driven by low-grade
energies, such as solar energy or waste heat. And compared with
other liquid desiccant air-conditioning systems reviewed in the
introduction, the system presented in this paper has some additional
advantages: making the best use of residential space exhaust air to
cool the process air, having the character of compact structure,
improving the indoor air quality. The effects of changing five key
variables on the performances of this novel liquid desiccant air-con-
ditioning system have been studied. Increasing the inlet solution
temperature in regenerator can improve the system’s performance,
but it is also restricted by the crystallization limit of desiccant
solution. The appropriate mass fluxes of air in the dehumidifier
and the regenerator should be accommodated to get this liquid des-
iccant system performance better. If the supply air temperature is re-
quired under 20 °C, the air-to-dehydrated desiccant mass flow rate
ratio m
a
/m
d
in the dehumidifier must be less than 2.0 [kg/kg]. When
the weak solution is regenerated adequately, it is only a waste of
energy if we continuously increase the mass flow rate of air into
the regenerator. The liquid desiccant system is suit for operating at
wide ranges of ambient air temperature and relative humidity
conditions.
References
[1] H. Factor, G.A. Grossman, A packed bed dehumidifier/regenerator for solar air-
conditioning with liquid desiccant, Solar Energy 24 (1980) 541–550.
[2] G. Scalabrin, G. Scaltriti, A liquid sorption–desorption system for air
conditioning with heat at lower temperature, Journal of Solar Energy
Engineering 112 (1990) 70–75.
[3] T.A. Ameel, K.G. Gee, B.D. Wood, Performance predictions of alternative low
cost absorbents for open-cycle absorption solar cooling, Solar Energy 54 (2)
(1995) 65–73.
[4] S. Jain, P.L. Dhar, S.C. Kaushik, Experimental studies on the dehumidifier and
regenerator of a liquid desiccant cooling system, Applied Thermal Engineering
20 (2000) 253–267.
[5] H.-M. Hellmann, G. Grossman, Simulation and analysis of an open-cycle
dehumidifier-evaporator-regenerator (DER) absorption chiller for low-grade
heat utilization, International Journal of Refrigeration 18 (3) (1995) 177–189.
[6] J.P. Pohl, H.-M. Hellmann, G. Grossman, Investigation and comparison of two
configurations of a novel open-cycle absorption chiller, International Journal of
Refrigeration 21 (2) (1998) 142–149.
[7] K. Gommed, G. Grossman, F. Ziegler, Experimental investigation of a LiCl-water
open absorption system for cooling and dehumidification, Journal of Solar
Energy Engineering 126 (2004) 710–715.
[8] K. Gommed, G. Grossman, A liquid desiccant system for solar cooling and
dehumidification, Journal of Solar Energy Engineering 126 (2004) 879–885.
[9] K. Gommed, G. Grossman, Experimental investigation of a liquid desiccant
system for solar cooling and dehumidification, Solar Energy 81 (2007) 131–
138.
[10] A.A. Kinsara, M.M. Elsayed, O.M. Al-Rabghi, Proposed energy-efficient air-
conditioning system using liquid desiccant, Applied Thermal Engineering 16
(10) (1996) 791–806.
[11] A.A. Kinsara, O.M. Al-Rabghi, M.M. Elsayed, Parametric study of an energy
efficient air conditioning system using liquid desiccant, Applied Thermal
Engineering 18 (5) (1997) 327–335.
[12] C.S.K. Ahmed, P. Gandhidasan, A.A. Al-Farayedhi, Simulation of a hybrid liquid
desiccant based air-conditioning system, Applied Thermal Engineering 17 (2)
(1997) 125–134.
[13] M.J. Peng, Exergy efficiency evaluation and analysis of indirect evaporative
cooling in plate heat exchangers, Master Thesis, College of Mechanical and
Automotive Engineering, Hunan University, Changsha, China 2005.
[14] A. Ertas, P. Gandhidasan, I. Kiris, E.E. Anderson, Experimental study on the
performance of a regeneration tower for various climatic conditions, Solar
Energy 53 (1) (1994) 125–130.
[15] S. Patnaik, T.G. Lenz, G.O.G. Lo
¯f, Performance studies for an experimental solar
open-cycle liquid desiccant air dehumidification system, Solar Energy 44 (3)
(1990) 123–135.
[16] C.Q. Ren, Y. Jiang, Y.P. Zhang, Simplified analysis of coupled heat and mass
transfer processes in packed bed liquid desiccant-air contact system, Solar
Energy 80 (2006) 121–131.
[17] X.C. Guo, T.S. Zhao, A parametric study of an indirect evaporative air cooler, Int.
Comm. Heat Mass Transfer 25 (2) (1998) 217–226.
[18] S.M. Yang, W.Q. Tao, Heat Transfer, third ed., Higher Education Press, Beijing,
China, 1998.
[19] A.A. Hasan, D.Y. Goswami, S. Vijayaraghavan, First and second law analysis of a
new power and refrigeration thermodynamic cycle using a solar heat source,
Solar Energy 73 (5) (2002) 385–393.
[20] C.S.K. Ahmed, P. Gandhidasan, S.M. Zubair, A.A. Al-Farayedhi, Exergy analysis
of a liquid-desiccant-based, hybrid air-conditioning system, Energy 23 (1)
(1998) 51–59.
[21] C.Q. Ren, N.P. Li, G.F. Tang, Principles of exergy analysis in HVAC and
evaluation of evaporative cooling schemes, Building and Environment 37
(2002) 1045–1055.
M. Tu et al. / Applied Thermal Engineering 29 (2009) 2417–2425 2425