You are on page 1of 7

Performance analysis of the single-stage absorption heat transformer using a new

working pair composed of ionic liquid and water
Xiaodong Zhang
*
, Dapeng Hu
Department of Chemical Machinery, Dalian University of Technology, 2 Linggong Road, Dalian 116024, PR China
a r t i c l e i n f o
Article history:
Received 13 September 2011
Accepted 3 November 2011
Available online 12 November 2011
Keywords:
Single-stage absorption heat transformer
Working pair
Ionic liquid
Coefficient of performance
Exergy efficiency
a b s t r a c t
The performance simulation of a single-stage absorption heat transformer using a new working pair
composed of ionic liquids, 1-ethyl-3-methylimidazolium dimethylphosphate, and water (H
2
O þ [EMIM]
[DMP]), was performed based on the thermodynamic properties of the new working pair and on the
mass and energy balance for each component of the system. In order to evaluate the new working pair,
the simulation results were compared with those of aqueous solution of lithium bromide (H
2
O þ LiBr),
Trifluoroethanol (TFE) þ tetraethylenglycol dimethylether (E181).
The results indicate that when generation, evaporation, condensing and absorption temperatures are
90

C, 90

C, 35

C and 130

C, the coefficients of performance of the single-stage absorption heat
transformer using H
2
O þ LiBr, H
2
O þ [EMIM][DMP] and TFE þ E181 as working pairs will reach 0.494,
0.481 and 0.458 respectively. And the corresponding exergy efficiency will reach 0.64, 0.62 and 0.59,
respectively. Meanwhile the available heat outputs for per unit mass of refrigerant are 2466 kJ/kg,
2344 kJ/kg and 311 kJ/kg, respectively.
The above excellent cycle performance together with the advantages of negligible vapor pressure, no
crystallization and more weak corrosion tendency to iron-steel materials may make the new working
pair better suited for the industrial absorption heat transformer.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
In order to reduce the CO
2
discharge and reuse large amounts of
industrial waste heat, absorption heat transformers (AHT) which
can upgrade waste heat fromlowtemperature level to a higher one
have been used in some industrial processes [1]. The available
temperature lift for single-stage absorption heat transformers
(STAHT) is usually in the range of 30

Ce40

C, while a higher
available temperature lift must be obtained by adopting two-stage
absorption heat transformer (TSAHT) or double absorption heat
transformer (DAHT).
The cycle performance of an AHT not only depends on its
configuration (for example, STAHT or DAHT) but also on thermo-
dynamic properties of workingpairs usuallycomposedof refrigerant
andabsorbent. The commonly-usedworkingpairs inAHTis aqueous
solution of lithium bromide (H
2
O þ LiBr). However, corrosion and
crystallization are its major disadvantages in industrial applications.
Therefore, seeking more advantageous working pairs with good
thermal stability,minimum corrosion and without crystallization
has become the research focus in the past two decades.
The organic working pairs such as Trifluoroethanol
(TFE) þ tetraethylenglycol dimethylether (E181), TFE þ N-methyl1-
2-pyrrolidone (NMP), etc are usually non-corrosive, completely
miscible and thermally stable. These excellent physical properties
have attracted many researchers to study the cycle performance of
absorption refrigerator, absorption heat pump (AHP) and AHT
adopting these organic working pairs [2e12].
Ionic liquids (ILs) are new types of environmental-friendly
solvent and remain liquid state at near or below room tempera-
ture. ILs are composed of the organic cation and inorganic anion.
They constitute a huge family of chemicals because there are many
types of organic cation and inorganic anion which can be converted
into ionic liquids. In the past decades, ILs have attracted consider-
able attention because of their unique properties, e.g. negligible
vapor pressure, non-flammability, good thermal stability, low
melting points, wide range of liquid state from room temperature
up to 200 or 300

C, and good solubility to many organic or inor-
ganic chemicals.
Because of their excellent properties, it is very possible that ILs
can be used as a new type of absorbent of refrigerants. With this
goal in mind, some researchers began to look for the suitable ILs
which can match common refrigerants and have excellent physical
and chemical properties. Kima and Shin et al. researched the
* Corresponding author. Tel.: þ86 411 84986171; fax: þ86 411 84986001.
E-mail address: zhangxd56@163.com (X. Zhang).
Contents lists available at SciVerse ScienceDirect
Applied Thermal Engineering
j ournal homepage: www. el sevi er. com/ l ocat e/ apt hermeng
1359-4311/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2011.11.006
Applied Thermal Engineering 37 (2012) 129e135
pressure and heat capacity of binary solution of TFE þ 1-butyl-3-
methylimidazolium bromide ([bmim][Br]) and TFE þ 1-butyl-3-
methylimidazolium tetrafluoroborate ([bmim][BF
4
]). They found
that these solutions were the negative deviation from Raoult’s Law
[13]. Shiflett and Yokozeki conducted extensive researches on the
solubility and diffusivity of carbon dioxide in ionic liquids [14]. They
also stuided the solubility of NH
3
in some ionic liquids [15]. Wang
and Zheng et al. measured and correlated vapor pressure of binary
solutions, water þ 1,3-Dimethylimidazolium Chloride and TFE þ 1-
Ethyl-3-methylimidazolium Tetrafluoroborate, which were taken
as working fluids of absorption refrigeration [16].
The authors measured and correlated the thermodynamic
properties of some binary solutions composed of water, ethanol,
methanol, and ILs such as 1-Ethyl-3-methylimidazolium ethyl-
sulfate, 1,3-dimethylimidazolium dimethylphosphate ([MMIM]
[DMP]), 1-ethyl-3-methylimidazoliumdimethylphosphate ([EMIM]
[DMP]), and 1-ethyl-3-methylimidazolium diethylphosphate
([EMIM][DEP]).They found that these binary solutions showed
strong negative deviation from Raoult’s Law and had significant
mixing heat. It is also shown that these ILs are thermo-stable until
220

C with the exception of 1-Ethyl-3-methylimidazolium ethyl-
sulfate and are entirely miscible with water, methanol and ethanol
[17e19].
Upto now the reports about the thermodynamic cycle perfor-
mance using these new working pairs containing ionic liquids are,
however, very limited. Yokozeki and Shiflett researched water
solubility in some ionic liquids and calculated the coefficients of
performance of absorption refrigeration under a given operation
condition [20]. Martín and Bermejo simulated the performance of
absorption refrigeration cycle using working pair of ionic
liquid þ supercritical CO
2
[21]. Liang and Zhao et al. simulated the
performance of absorption refrigeration cycle using
methanol þ [MMIM][DMP] as working pair [22]. The authors also
simulated the performance of absorption refrigeration cycle using
H
2
O þ [EMIM][DMP] [23].
However, the cycle performance analysis on the AHT using
working pairs containing ILs has not been found in these
references.
The aim of this research is to further prove if the new working
pair, H
2
O þ [EMIM][DMP], has other excellent cycle performances
in AHT cycle besides its excellent physical and chemical properties.
So it is necessary to simulate and analyze the thermodynamic cycle
performance of a STAHT based on the thermodynamic properties of
binary solutions, H
2
O þ [EMIM][DMP], obtained in our laboratory
[19]. In order to evaluate the present working pair containing ILs,
the simulation results have also been compared with those using
H
2
O þ LiBr and TFE þ E181.
2. The thermodynamic properties of H
2
O D [EMIM][DMP]
The thermodynamic properties of binary solution H
2
Oþ[EMIM]
[DMP] have been reported in our previous studies [19,23], here
some thermodynamic properties related to the performance
simulation are briefly given as follows:
2.1. The pressure of solution
Because of non-volatilization of ionic liquid, [EMIM][DMP], the
vapor phase is only consisted of water, the pressure of this systemis
p ¼ g
1
x
1
p
s
1
(1)
Where, g
1
is the activity coefficient of water, x
1
is the mole fraction
of water in solution, p
s
1
is the saturation vapor pressure of pure
water.
The activity coefficients of water in the solution H
2
O þ [EMIM]
[DMP] is given in the form of non-random two liquid model (NRTL)
as follows.
lng
1
¼ x
2
2
"
s
21

G
21
x
1
þ x
2
G
21

2
þ
s
12
G
12
ðx
2
þ x
1
G
12
Þ
2
#
(2)
G
12
¼ expðÀas
12
Þ; G
21
¼ expðÀas
21
Þ (3)
s
12
¼
Dg
1
RT
; s
21
¼
Dg
2
RT
(4)
Where, a ¼ 0.6004 is the non-random parameter of the NRTL
model. Dg
1
and Dg
2
are both binary interaction parameters and
expressed as follows:
Dg
i
¼ a
i
þ b
i
T þ c
i
T
2
ði ¼ 1; 2Þ (5)
Nomenclature
AHT absorption heat transformers
[bmim][Br] 1-butyl-3-methylimidazolium bromide
[bmim][BF
4
] 1-butyl-3-methylimidazolium tetrafluoroborate
COP coefficient of performance
ECOP exergy efficiency
E181 tetraethylenglycol dimethylether
[EMIM][DMP] 1-ethyl-3-methylimidazolium
dimethylphosphate
[EMIM][DEP] 1-ethyl-3-methylimidazolium diethylphosphate
F flow rate ratio
h specific enthalpy (kJ/kg)
m mass flow rate (kg /s)
[MMIM][DMP] 1,3-dimethylimidazolium dimethylphosphate
NMP N-methyl1-2-pyrrolidone
p pressure ( MPa)
P1 solution pump
P2 refrigerant pump
q¼Q
A
/m
1
available heat output per unit mass of refrigerant (kJ/
kg)
Q heat load (kW)
SHE solution heat exchanger
STAHT single-stage absorption heat transformers
t temperature,

C
T temperature, K
TFE Trifluoroethanol
v value
x mass fraction of absorbent in solution
Dx ¼ x
s
À x
w
concentration difference between strong and weak
solutions
Subscripts
A absorber
C condenser
E evaporator
G generator
s strong solution
w weak solution
1,2,3.10 the series number of streams
X. Zhang, D. Hu / Applied Thermal Engineering 37 (2012) 129e135 130
Where a
i
, b
i
, and c
i
are all regression parameters and given in
Table 1.
The saturated vapor pressure of pure waterp
s
1
, can be calculated
by Antoine Equation (6).
lnðp
s
=kPaÞ ¼ 16:28837 þ3816:4=½À46:13 þ ðT=Kފ (6)
2.2. Enthalpy of solution
Enthalpy of the solution H
2
O þ [EMIM][DMP] at T and at IL mole
fraction x
2
can be calculated by the Equation (7).
H ¼ H
298
þ
Z
T
298
C
P
dT (7)
Where, H
298
(kJ/kmol) is the specific enthalpy of solution with IL
mole fraction x
2
at 298 K, and is shown in Equation (8).
H
298
¼ H
E
298
þ x
2
Z
298
273
C
P;½EMIMŠ½DMPŠ
dT þ x
1
Z
298
273
C
P;½H2OŠ
dT (8)
Where, H
E
298
is the excess enthalpy of solution, in kJ/kmol and
calculated by Equation (9), C
P,[EMIM][DMP]
and C
P,H2O
are the specific
heat capacities of [EMIM][DMP] and H
2
O, in kJ/kmol K, respectively.
H
E
298
=x
1
x
2
¼
X
4
1
A
i
ð1 À2x
2
Þ
iÀ1
(9)
Where, A
i
(i ¼ 1, 2, 3, 4) are the regression parameters and given in
Table 2, x
1
and x
2
are the molar fraction of water and IL respectively.
2.3. Specific heat capacity of solution
Specific heat capacity of solution, C
P
, is given as follows,
C
p
¼ C
p0
þ a  T (10)
Where,
C
p0
¼
X
5
i ¼1
A
i
x
iÀ1
2
(11)
a ¼
X
5
i ¼1
B
i
x
iÀ1
2
(12)
Where, C
p
is the heat capacity in kJ/kg K, A
i
and B
i
are both the
regressive parameters and given in Table 3, T is the absolute
temperature in K, and x
2
is the molar fraction of [EMIM][DMP] in
solution.
The thermodynamic properties of working pairs TFE þE181 and
H
2
O þ LiBr were given in references [4,5,24e26].
3. The performance simulation
The configuration of a STAHT and its corresponding absorption
cycle are schematically shown in Fig. 1 and Fig. 2. It is mainly
composed of a generator, a condenser, an evaporator, an absorber
and a solution heat exchange or the so called economizer.
In order to simulate the thermodynamic cycle performance of
STAHT using H
2
O þ [EMIM][DMP] as working pairs, the following
assumptions are made.
(1) The analysis is carried out under steady state conditions;
(2) The solutions are saturated while leaving the generator and the
absorber;
(3) The condensate is saturated and not subcooled while leaving
the condenser;
(4) The thermal and pressure losses during the whole process can
be neglected;
(5) The expanding process in the throttling valves is isenthalpic;
(6) The minimum temperature difference between the cold and
hot streams in solution heat exchange is 5

C;
(7) The mechanical energy consumed by the pumps can be
neglected;
It is necessary to get the mass and energy balance equations for
each component and get parameters for main state points in the
cycle in order to simulate the thermodynamic cycle performance of
STAHT with this new working pair containing ILs. Generator
m
7
À m
1
À m
8
¼ 0 (13)
m
7
x
7
¼ m
8
x
8
(14)
Q
G
þ m
7
h
7
À m
8
h
8
À m
1
h
1
¼ 0 (15)
For state point 8.
Table 1
Parameters in the Equation (5).
a
1
b
1
c
1
a
2
b
2
c
2
À4.0495 Â 10
5
2.1839 Â 10
3
À2.810 À7.5891 Â 10
3
10.215 À0.0356
Table 2
Parameters in Equation (9).
A
1
A
2
A
3
A
4
À2.3379 Â 10
4
À1.1000 Â 10
4
À7.9100 Â 10
3
À9.0166 Â 10
3
Table 3
Parameters in Equations (11) and (12).
A
1
À8.4310 Â 10
2
A
2
À2.0572 Â 10
4
A
3
1.3885 Â 10
4
A
4
3.1382 Â 10
4
A
5
À2.6643 Â 10
4
B
1
14.7825
B
2
26.3706
B
3
53.2389
B
4
À207.8705
B
5
127.5218
E
C
Q
E
Q
A
Q
G
Q
C
1
7 8
2
3
4
5
A
10
SHE
6 9
G
P1
P2
v
Fig. 1. Schematic diagram of absorption heat transformer A-absorber, C-condenser, E-
evaporator, G-generator, SHE-heat exchanger, P1,P2-pump, v-valve.
X. Zhang, D. Hu / Applied Thermal Engineering 37 (2012) 129e135 131
p
8
¼ pðt
8
; x
8
Þ (16)
where t
8
¼ t
G
and p
8
¼ p
c
.
Condenser
m
1
¼ m
2
¼ m
3
¼ 1 kg=s (17)
(Mass reference for calculation)
m
1
h
1
À m
2
h
2
À Q
C
¼ 0 (18)
For state point 2
p
2
¼ p
c
¼ p
s
ðt
C
Þ (19)
Evaporator
m
3
¼ m
4
(20)
m
4
h
4
À m
3
h
3
À Q
E
¼ 0 (21)
For state point 4.
p
4
¼ p
E
¼ p
s
ðt
E
Þ (22)
Absorber
m
4
þ m
10
Àm
5
¼ 0 (23)
m
5
x
5
¼ m
10
x
10
(24)
m
4
h
4
þ m
10
h
10
À m
5
h
5
À Q
A
¼ 0 (25)
For state point 5
p
5
¼ pðt
5
; x
5
Þ (26)
where t
5
¼ t
A
and p
5
¼ p
E
.
Heat exchanger
m
5
h
5
þ m
9
h
9
À m
6
h
6
À m
10
h
10
¼ 0 (27)
t
5
À t
10
¼ 5 (28)
Here the minimum temperature difference between the cold
and hot streams in solution heat exchange is located in the hot end
of the solution heat exchange.Where m
i
, h
i
, x
i
(i ¼1,2,3,.10) are the
mass flow rate, specific enthalpy and mass fraction of absorbent,
[EMIM][DMP], of each stream respectively. p
s
(t) is saturated pres-
sure of refrigerant H
2
O, the p(t,x) are saturated pressure of working
solution H
2
Oþ[EMIM][DMP], which are given by Formulas (1)e(6).
Usually the temperatures of waste heat for driving an absorption
heat transformer, the cooling water temperature and the output
temperature of an absorption heat transformer are known in
applications, so the generation, evaporation condensation and
absorption temperatures t
G
, t
E,
t
C
and t
A
are known and taken as
independent variables in present research. Based on the above
assumptions and the mass, energy balance equations and phase
equilibrium Equations (13)e(28) for each component or for main
state points in the cycle, the parameters for each stream such as t
i
,
p
i
, x
i
and m
i
can be obtained. The heat loads of generator, absorber,
evaporator and condenser, the concentration of solution, concen-
tration deference between strong and weak solutions and flow rate
ratio are also obtained.
P
r
e
s
s
u
r
e


t
C
t
G
=t
E
t
A
Temperature
p
C
p
E
x
S
x
w H
2
O
4
10 5
8 7 2
Fig. 2. The thermodynamic cycle in absorption heat transformer.
Input t
E
, t
G
, t
C,
t
A
t
2
= t
C
, t
4
= t
E
, t
8
= t
G
t
5
= t
A
p
2
= p
s
(t
2
); p
4
=p
s
(t
4
), Eqs.(6), (19) ,(22)
p
2
=p
7
=p
8 ,
p
4
= p
5
=p
10
assumptions (4)
p
8
=p (t
8
,x
8
) Eqs. (16) and (1-6),
x
8
=x
s
can be obtained from above eqs by
Newtonian iteration approach, similarly x
5
=x
w
can also be obtained from Eqs. (26) and (1-6).
h
i
=h(t
i
) (i=1-10) Eqs. (7-12)
Q
i
(i=G, C, E, A) Eqs. (15, 18, 21, 25)
Calculating COP , ECOP, ∆x, F and q
start
Outputting COP, ECOP, ∆x ,F and q
end
based on Eqs. (29-34)
Fig. 3. Flow-chart for simulation.
X. Zhang, D. Hu / Applied Thermal Engineering 37 (2012) 129e135 132
The coefficient of performance, COP, is defined as the ratio of
available heat output of the absorber to the total heat inputs of the
system. Similarly the exergetic efficiency, ECOP, based onthe second
law of thermodynamics is defined as the ratio of the available
energy output of the absorber to the total available energy input of
the system. The above two coefficients are defined as follows as well
as the flowrate ratio F, concentration difference between strong and
weak solutions, Dx, and gross temperature lift, Dt.
COP ¼
Q
A
Q
G
þ Q
E
(29)
ECOP ¼
Q
A
ð1ÀT
0
=T
A
Þ
Q
E
ð1ÀT
0
=T
E
Þ þQ
G
ð1ÀT
0
=T
G
Þ
¼ COP
ð1ÀT
0
=T
A
Þ
ð1ÀT
0
=T
G
Þ
(30)
Where the T
E
and T
G
are assumed to be equal in present study.
Dx ¼ x
8
À x
5
¼ x
s
À x
w
(31)
D
t
¼ t
A
À t
E
(32)
F ¼
m
7
m
1
¼
x
s
Dx
(33)
q ¼
Q
A
m
1
(34)
Where T
0
, T
E
, and T
G
are environmental, evaporation and genera-
tion temperature, respectively.
Fig. 3 showed the flow-chart for simulation, in which the main
steps for simulation process and corresponding equations were
given.
4. Results and discussion
The simulating results for STAHT using the new working pair,
H
2
O þ [EMIM][DMP],are shown in Figs. 4e9.
From Fig. 4 it can be seen that the coefficients of performance of
aSTAHTfor threeworkingpairs will decreaseat different ratewiththe
increaseof t
A
. Amongthreeworkingpairs whengross temperaturelift
is not beyond 50

C, the COP is basically unchanged for H
2
O þ LiBr.
However, it decreases very rapidly for TFE þ E181, compared to
a moderate drop for H
2
O þ [EMIM][DMP]. As for the magnitude of
COP, they followed the order of H
2
O þ LiBr > H
2
O þ [EMIM]
[DMP] >TFEþE181. For example, whencondensingtemperature and
gross temperature lift are 35

C and 40

C, the coefficients of perfor-
manceare 0.494, 0.481and0.458for H
2
OþLiBr, H
2
Oþ[EMIM][DMP]
and TFE þ E181, respectively. It is also found that the lower the
condensing temperature is, the higher the COP or available temper-
ature lift will be. This is why the AHT is more efficient in winter.
Fig. 5 shows the variation of COP with the absorption temper-
ature t
A
under different generation or evaporation temperature
conditions. As shown in Fig. 5 the higher the generation or evap-
oration temperature is, the higher the absorption temperature and
corresponding gross temperature lift got. It is still consistent here
that the COP follows the order of H
2
O þ LiBr > H
2
O þ [EMIM]
[DMP] > TFE þ E181 under different generation or evaporation
temperature conditions.
0
0.1
0.2
0.3
0.4
0.5
0.6
100 110 120 130 140 150 160 170
t
A
/
C
O
P
H2O+[EMIM][DMP]
H2O+LiBr
TFE/E181
t G=t E=90
tc=30
35 40
Fig. 4. Effects of t
A
on COP for three working pairs.
0
0.1
0.2
0.3
0.4
0.5
0.6
90 100 110 120 130 140 150 160 170 180
t
A
C
O
P
H2O+[EMIM][DEP]
H2O+LiBr
TFE+E181
t C=35
t G=t E=70 80
90
Fig. 5. Effect of t
G
¼ t
E
on COP for three working pairs.
0.2
0.4
0.6
0.8
100 120 140 160 180
t A/
E
C
O
P
H2O+[EMIM][DMP]
H2O+LiBr
TFE+E181
t G=t E=90
40
t c=30
35
Fig. 6. Effects of t
A
on ECOP for three working pairs.
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
100 110 120 130 140 150 160 170 180
t A/
x
/
(
w
t
,
a
b
s
o
r
b
e
n
t
)
H2O+[EMIM][DEP]
H2O+LiBr
TFE+E181
xs
xw
t G=t E=90
t C=35
Fig. 7. Effects of t
A
on x
s
and x
w
for three working pairs.
X. Zhang, D. Hu / Applied Thermal Engineering 37 (2012) 129e135 133
Fig. 6 shows the variation of ECOP with the absorption tempera-
ture t
A
. It is revealed that the effects of t
E
on COP and ECOP differ
significantly. The COP decreases with the increase of t
A,
while ECOP
firstly increases to a maximum and then decreases rapidly. This can
beexplainedby thedefinitionFormula (30) listedabove, inwhichthe
ECOP equals tothe COPmultipliedby the factor of (1 ÀT
0
/T
A
)/(1 ÀT
0
/
T
G
). When COP decreases, the factor will however increase with the
increase of t
A
. In fact ECOP based on the second law of thermody-
namics not only takes account of the amount of available heat output
but also its temperature level, namely quality of the output heat. The
COP only represents heat output, while the ECOP represents not only
thequantitybut alsothequalityof heat output. Becauseof (1ÀT
0
/T
A
)/
(1 À T
0
/T
G
) > 1 for AHT cycle, the ECOP is larger than COP.
Fig. 7 shows the concentration variation of strong and weak
solutions with absorbing temperature t
A
. The strong solution
leaving the generator has a concentration x
s,
which enriches
absorbent more than the weak solution, so it can absorbs refrig-
erant vapor easier. When generating and condensing temperature
maintain unchanged, the x
s
does not vary with the t
A
, but the x
w
does increase with the t
A
. The higher the absorbing temperature t
A
or gross temperature lift Dt is, the denser the weak solution is. This
leads to the decrease of concentration difference Dx ¼ x
s
À x
w
or to
the increase of flow rate ratio F and mechanical power losses. As
shown in Fig. 6 the concentrations of strong solution x
s
are over 0.9
(mass fraction,[EMIM][DEP] or E181) for H
2
O þ [EMIM][DEP] and
TFE þ E181, while it is 0.645 (mass fraction, LiBr) for the common
working pair, H
2
O þ LiBr, under the given operation conditions.
Fig. 8 shows the variation of solution concentration difference
Dx ¼x
s
Àx
w
, and flowrate ratio F with absorbing temperature t
A
. As
shown in Fig. 8, the solution concentration differences of three
working pairs decrease with the increase of t
A
, but the flow rate
ratios vary in an opposite way. When generation, evaporation and
condensing temperature are maintained to be constant, the solu-
tion concentration difference and the available output temperature
of absorber will only vary with the flow rate ratio, so the flow rate
ratio is an important and easily controllable operation parameter.
Larger flow rate ratio also results in higher output temperature of
the absorber and more mechanical power loss. Under the same
operation conditions the flow rate ratio is in the order of
H
2
O þ [EMIM][DEP] > H
2
O þ LiBr > TFE þ E181.
Usually the flowrate ratio is about 12 for working pair H
2
OþLiBr.
According to the Formula (20) above, the concentration difference,
Dx ¼ x
s
/F, is about 0.05 (wt, LiBr).However, for working pair
H
2
O þ [EMIM][DEP] the concentration of strong solution is over 0.9
(wt., [EMIM][DEP]), so the corresponding concentration difference is
up to 0.08 (wt., [EMIM][DEP]) under the same flow rate ratio condi-
tion. For working pair TFE þ E181 a similar concentration difference
of 0.08 is found. The larger the concentration difference is, the larger
the driving force for mass transfer in generator or absorber is. This
will be especially beneficial for the absorbing process inthe absorber.
Fig. 9 shows the variation of available heat output from the
absorber for per unit mass of refrigerant (as defined as Formula (21)
above) with absorbing temperature t
A
. It should be noted that the
available heat output of per unit mass of refrigerant are similar for
working pair H
2
Oþ[EMIM][DEP] andH
2
OþLiBr, whichare 2466 kJ/
kg and 2344 kJ/kg respectively. However, for the working pair
TFE þ E181 the number is 311 kJ/kg under the same conditions of
t
G
¼t
E
¼90

C,tc ¼35

Candt
A
¼130

C. Themainreasonis that the
evaporation heat of TFE is very lowcompared with water. Under the
same operation conditions, the amount of refrigerant, TFE, cycled in
theSTAHTwill be 7times of that of water, whichleads tothe increase
of mechanical power losses and larger size of pipes and pumps.
5. Conclusion
The conclusions of this research are summarized as follows:
1. This new working pair containing ILs has a similar cycle
performance as that of H
2
O þ LiBr. The coefficient of perfor-
mance and exergy efficiency of H
2
O þ [EMIM][DMP] is slightly
less than those of H
2
O þ LiBr but they are still larger than those
of TFE þ E181. When condensing temperature and gross
temperature lift Dt are 35

C and 40

C, the coefficients of
performance reaches 0.494, 0.481 and 0.458 for H
2
O þ LiBr,
H
2
O þ [EMIM][DMP] and TFE þ E181, respectively, while the
corresponding exergy efficiency is 0.64, 0.62 and 0.59 under
the same operating conditions.
2. The concentration of strong solution is over 0.9 for
H
2
O þ [EMIM][DMP] and TFE þ E181, while it is only 0.64 for
H
2
O þ LiBr. Under the same flow rate ratio condition the
concentration difference for H
2
O þ [EMIM][DMP] and
TFE þE181 is larger than that of H
2
OþLiBr, which will improve
absorbing process in the absorber
3. The available heat output of per unit mass of refrigerant for
H
2
O þ [EMIM][DMP] is 2344 kJ/kg and slightly less than
2466 kJ/kg for H
2
O þ LiBr, but it is much larger than 311 kJ/kg
for TFE þ E181. So under the same available heat output
conditions, the amount of water cycled in the STAHT will be
much less than that of TFE. This will reduce the mechanical
power losses and the size of pipes and pumps.
Finallybecauseof the excellent cycle performances together with
the advantages of negligible vapor pressure, no crystallization and
more weak corrosion tendency to iron-steel materials than aqueous
solutionof lithiumbromide, the newworking pair researchedinthis
paper has a great potential tobesuccessfullyappliedtotheindustrial
AHT with a significantly better performance.
0
0.1
0.2
0.3
0.4
0.5
100 110 120 130 140 150 160 170
t
A
/

x
/
(
w
t
,
a
b
s
o
r
b
e
n
t
)
0
5
10
15
20
25
30
F
H2O+[EMIM][DEP]
H2O+LiBr
TFE+E181
t G=t E=90
tc=35
Fig. 8. Effects of t
A
on Dx and F for three working pairs.
0
500
1000
1500
2000
2500
3000
100 110 120 130 140 150 160
t
A
/
q
/
(
k
J
/
k
g
)
H2O+[EMIM][DEP]
H2O+LiBr
TFE+E181
t G=t E=90
t c=35
Fig. 9. Effects of t
A
on q for three working pairs.
X. Zhang, D. Hu / Applied Thermal Engineering 37 (2012) 129e135 134
Acknowledgement
The authors are grateful for the financial support from National
Nature Science Foundation Committee of China (No: 51076021, No:
50876014) to this research project.
References
[1] H. Ilhami, K. Bener, Single stage and double absorption heat transformers in an
industrial application, Int. J. Energ. Res. 33 (2009) 787e798.
[2] M. Narodowslawsky, G.O.F. Moser, New working pairs for medium and high
temperature industrial absorption heat pumps, Heat Recov. Syst.&CHP 8
(1988) 459e468.
[3] P.F. Bidoz, T. Kashiwagi, Simulation of a new absorption temperature amplifier
using a TFE-E181 pair for low-level temperature upgrading, ASHRAE Trans.
Res. 97 (1991) 156e162.
[4] K. Stephan, R. Hengerer, Heat transformation with the ternary working fluid
TFE-H
2
O-E181, Int. J. Refrig. 16 (1993) 120e128.
[5] A. Coronas, M. Valles, S.K. Chaudhari, K.R. Patil, Absorption heat pump with
the TFE-E181 and TFE-H
2
O-E181, Appl. Therm. Eng. 16 (1996) 335e345.
[6] M. Bourouis, M. Nogues, D. Boer, A. Coronas, Industrial heat recovery by
absorption/compression heat pump using TFE-HO-TEGDME working mixture,
Appl. Therm. Eng. 20 (2000) 355e369.
[7] A. Genssle, K. Stephan, Analysis of the process characteristics of an absorption
heat transformer with compact heat exchangers and the mixture TFE-E181,
Int. J. Therm. Sci. 39 (2000) 30e38.
[8] J. Yin, L. Shi, Performance analysis of an absorption heat transformer with
different working pairs, J. Refrig. (Chinese) 3 (2000) 19e24.
[9] X. Wang, L. Shi, J. Yin, M.S. Zhu, A two-stage heat transformer with H
2
O/LiBr
for the first stage and 2,2,2-trifluoroethanol(TFE)/N-methyl1-2-
pyrrolidone(NMP)for the second stage, Appl. Energ. 71 (2002) 235e249.
[10] Z.C. Zhao, F.W. Zhou, S.P. Li, Analysis of thermodynamic performance of
double absorption heat transformer using TFE-E181 as working fluids,
J. Dalian Univ. Technol. 43 (2003) 604e608.
[11] Z.C. Zhao, F.W. Zhou, S.P. Li, Analysis of thermodynamic cycle of Type II high
temperature absorption heat pump using TFE-E181 as working fluids, J. Dalian
Univ. Technol. 44 (2004) 651e656.
[12] Z.C. Zhao, X.D. Zhang, X.H. Ma, Thermodynamic performance of a double-
effect absorption heat-transformer using TFE/E181 as the working fluid,
Appl. Energ. 82 (2005) 107e116.
[13] K.S. Kima, B.K. Shin, H. Lee, F. Ziegler, Refractive index and heat capacity of 1-
butyl-3-methylimidazolium bromide and 1-butyl-3-methylimidazolium tet-
rafluoroborate, and vapor pressure of binary systems for 1-butyl-3-
methylimidazolium bromide þ trifluoroethanol and 1-butyl-3-
methylimidazolium tetrafluoroborate þ trifluoroethanol, Fluid Phase Equi-
libr. 218 (2004) 215e220.
[14] M.B. Shiflett, A. Yokozeki, Solubilities and Diffusivities of carbon dioxide in
ionic liquids: [bmim][PF6] and [bmim][BF4], Ind. Eng. Chem. Res. 44 (2005)
4453e4464.
[15] A. Yokozeki, M.B. Shiflett, Vaporeliquid equilibria of ammonia þ ionic liquid
mixtures, Appl. Energ 84 (2007) 1258e1273.
[16] J.Z. Wang, D.X. Zheng, L.H. Fan, D. Li, Vapor pressure measurement for the
water þ 1,3-dimethylimidazolium chloride system and 2,2,2-
trifluoroethanol þ 1-ethyl-3-methylimidazolium tetrafluroborate, J. Chem.
Eng. Data 55 (2010) 2128e2132.
[17] G.L. Zuo, Z.C. Zhao, S.H. Yan, X.D. Zhang, Thermodynamic Properties of a new
working pair: 1-Ethyl-3-methylimidazolium Ethylsulfate and Water, Chem.
Eng. J. 156 (2010) 613e617.
[18] Z.B. He, Z.C. Zhao, X.D. Zhang, Thermodynamic properties of new heat pump
working pairs: 1,3-dimethylimidazolium dimethylphosphate and water,
ethanol and methanol, Fluid Phase Equilibr. 298 (2010) 83e91.
[19] J. Ren, Z.C. Zhao, X.D. Zhang, Vapor pressure, excess enthalpies, and specific
heat capacities of the binary working pairs containing the ionic liquid 1-ethyl-
3-methylimidazolium dimethylphosphate, J. Chem. Thermodyn. 43 (2011)
576e583.
[20] A. Yokozeki, M.B. Shiflett, Water solubility in ionic liquids and application to
absorption cycles, Ind. Eng. Chem.Res. 49 (2010) 9496e9503.
[21] A. Martín, M.D. Bermejo, Thermodynamic analysis of absorption refrigeration
cycles using ionic liquid þ supercritical CO2 pairs, J. Supercrit. Fluids 55 (2010)
852e859.
[22] S.Q. Liang, J. Zhao, L. Wang, X.L. Huai, Absorption refrigeration cycle utilizing
a new working pair of ionic liquid type, J. Eng. Thermophysics (Chinese) 10
(2010) 1627e1630.
[23] X.D. Zhang, D.P. Hu, Performance simulation of the absorption refrigerator
using ionic liquid 1-ethyl-3-methylimidazolium dimethylphosphate and
water as working pair, Appl. Therm. Eng. 31 (2011) 3316e3321.
[24] M.R. Patterson, H.P. Blanco, Numerical fits of properties of lithium-bromide
water solutions, ASHRAE Trans. 94 (1988) 2059e2077.
[25] M.M. Talbi, B. Agnew, Exergy analysis: an absorption chiller using lithium
bromide and water as the working fluids, Appl. Therm. Eng. 20 (2000)
619e630.
[26] D.W. Sun, Thermodynamic design data and optimum design maps for
absorption refrigeration systems, Appl. Therm. Eng. 17 (1997) 211e221.
X. Zhang, D. Hu / Applied Thermal Engineering 37 (2012) 129e135 135