Available via license: CC BY 4.0
Content may be subject to copyright.
INTERNATIONAL JOURNAL OF
ENERGY AND ENVIRONMENT
Volume 1, Issue 3, 2010 pp.459-466
Journal homepage: www.IJEE.IEEFoundation.org
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
Experimental study of operation performance of a low
power thermoelectric cooling dehumidifier
Wang Huajun, Qi Chengying
School of Energy and Environment Engineering, Hebei University of Technology, Tianjin, China.
Abstract
The present work was performed to apply thermoelectric technology to a low power dehumidifying
device as an alternative to the conventional vapor-compression refrigeration systems. The experimental
prototype of a small-scale thermoelectric dehumidifier (TED) with rectangular cooling fins was built and
its operation performance was studied experimentally. The results showed that the TED experienced two
typical thermodynamic processes including the cooling dehumidification and the isothermal
dehumidification, where the latter was dominated. It was found that there existed a peak during the
variation of the average coefficient of performance (COP) as a function of the input power of the
thermoelectric module. Under the present experimental conditions, the COP of the TED reached the
maximum of 0.32 and the corresponding dehumidifying rate was 0.0097 g/min, when the input power
was kept at 6.0 W. The rapid elimination of condensed liquid-drops on the cooling fins amounted on the
thermoelectric module is a major approach to improving the operation performance of the TED.
Copyright © 2010 International Energy and Environment Foundation - All rights reserved.
Keywords: Cooling dehumidification, Thermoelectric module, COP, Dehumidification rate.
1. Introduction
For the indoor thermal comfort of buildings, the relative humidity is a key control parameter. At present,
there exist a number of dehumidifying devices based on the traditional vapor-compression technology.
Though these devices are widely applied, they still have many problems to be solved, such as high noise
level, strong compressor vibration, and excessive weight and so on [1-2]. In fact, what’s more important
is the extensive use of CFCs or HCFCs, which has a great negative impact on the present crisis of energy
and environment [3].Under this background, it is required to seek for other clean dehumidifying
approaches. Thermoelectric cooling technology is just the case in point, which is based on the Peltier
effect of semiconductor materials [4]. At present, the thermoelectric cooling technology has been
developed rapidly, as an alternative to the conventional systems based on the vapor-compression
refrigeration cycles.
For the thermoelectric dehumidification, however, only a few studies were performed. In fact,
thermoelectric dehumidifiers have a promising potential in the situations where both a lower power and a
smaller space are required. Such applications can be seen in the space crafts, submarines, robots,
tunnels,
and industrial control cabinets and so on [5-6]. In previous studies, Chen [7] presented a TED with an
improved fin heat exchanger, reaching a COP ranging from 0.12 to 0.29. Sumrit [8] tested the
performance of a TED in a residential apartment, and obtained a COP of 0.27. In fact, except the
difference in the thermoelectric module materials, the better performance of the TED depends mainly on
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
460
an optimal layout of heat transfer surfaces. However, too much complex controls on heat transfer usually
mean a higher construction cost of TEDs, which may reduce the feasibility in practical applications. The
purpose of this work was to test the operation performance of the TED, and to obtain the relationship
between the COP and the input power, which is helpful for the optimal design of similar TEDs.
2. Experimental system and procedure
2.1 Experimental setup
Figure 1 shows the experimental setup of a small-scale TED. It mainly consisted of a closed cabinet, a
thermoelectric module, a DC power meter, a humidifier, a humidiometer, temperature sensors, and a data
acquisition system. The cabinet made of organic glass plates was used to simulate a small room. The
cabinet was cube-shaped, with 400 mm in length, 400 mm in width, and 400 mm in height. A TEC-
12705 type thermoelectric module made of BiTe was used to provide a cold source for dehumidification.
The module was 40 mm in length, 40mm in width, and 4mm in thickness, and its maximum working
voltage and current were 15 V and 4 A. In order to enhance heat transfer, two rectangular fin heat
exchangers were amounted on the heat and cold side of the thermoelectric module, with an extended area
of about 91 cm
2
and 128 cm
2
, respectively. A 2.4W axial fan for forced convection was amounted on the
fin heat exchanger of the heat side of the TED. In order to eliminate the effect of contact resistances as
possible, a thin layer of silicone grease with silver was covered on the heat-transfer surfaces of the TED.
An adjustable WYJ-20 type DC power meter was used to electrically driven the thermoelectric module.
Its maximum voltage and current output were 20 V and 10 A, respectively.
The relative humidity inside the cabinet was measured by a HT-3005A type humidiometer with ± 1%
accuracy installed at the middle of the cabinet. The temperatures of the cabinet and heat/cold fin heat
exchangers were measured by copper- constantan thermocouples with ± 0.1
o
C accuracy. The temperature
data were recorded by a HP 34970A type recorder. All measuring sensors were calibrated carefully
during the experimental preparation. The ambient atmospheric pressure was measured by a mercury
manometer. According to the Chinese Code for Design of Heating Ventilation and Air Conditioning (GB
50019-2003), the allowable indoor relative humidity should range from 40 % to 65 % in summer and
from 40 % to 60 % in winter. Therefore, in this experiment, the lower limit of the relative humidity was
set as 40 ±1 %, and the upper limit was set as 98 ±1 % to simulate the situation with a high humidity
environment of buildings. Before the experiment, the cabinet was humidified to a high relative humidity
by a humidifier. After 10 min, the DC power supply of the thermoelectric module was connected. During
the dehumidifying process, the variations of the temperature and relative humidity were recorded at an
interval of 5 min.
humidifier
Data recorder
Computer
Ambient monitoring meters
Camera
Cabinet
Humidity sensor
Cold-side fins
Thermoelectric module
Cooling fan
DC
p
ower mete
r
Heat-side fins
Thermocouple wires
Figure 1. Schematic diagram of the experimental setup of a TED
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
461
2.2 Experimental data reduction
In order to evaluate the performance of the TED, the COP is defined as
c
TE F
Q
COP
PP
=
+
(1)
where P
TE
is the power of the thermoelectric module (W), and P
F
is the power of the cooling fan(W).
The total heat-transfer rate (Q
c
) of the fin heat exchanger on the cold side of the TED is calculated as
()
cccoc wc
QhAtt mH=−+ (2)
where h
c
is the coefficient of convective heat transfer (W/m
2
K), A
c
is the heat transfer area (m
2
), t
o
is the
temperature at the middle of the cabinet (
o
C), t
c
is the average temperature of cold fins (
o
C), and H
c
is the
latent heat of condensation (J/kgK).
The dehumidifying rate (m
w
, kg/s) is calculated as
()
12a
w
m
m
T
φ
φ
−
= (3)
where m
a
is the mass of the wet air inside the cabinet (kg), T is the dehumidifying period (s), Φ
1
and Φ
2
are the relative humidity before and after dehumidification (%).
According to the previous study by Bejan [9], the coefficient of convective heat transfer (h
c
, W/m
2
K)
between adjacent fins and the ambient can be determined by
0.25
0.517
c
hRa
H
λ
= (4)
where λ is the thermal conductivity of air (W/mK), H is the height of fins (m), and Ra is the
dimensionless Rayleigh number, and its definition and calculations can be seen in the previous studies by
Bejan and Cengel [9-10].
3. Results and discussion
3.1 The analysis of the dehumidifying process of the TED
Figure 2 shows a typical dehumidifying process of the TED, when the input power of the thermoelectric
module is 5.5 W. It can be seen that, except the first 10 min, the average temperature of heat-side fins as
well as the cabinet kept relatively steady during the dehumidifying process. The average temperature of
cold-side fins appeared a slowly decreasing tendency, and finally reached a steady state. It was also
found that the relative humidity experienced two accelerating stages (i.e.98-90% and 70-50%) and two
decelerating stages (i.e.90-70% and 50-40%). This phenomenon is related with the condensing behavior
of cold-side fins, which is discussed in the later section.
The variation of the relative humidity shown in Figure 2 can also been plotted in the enthalpy-humidity
diagram. As shown in Figure 3, the TED experienced two typical thermodynamic processes, including
the cooling dehumidification and the isothermal dehumidification. During the cooling dehumidifying
process from A to B, the variation of the relative humidity was weak, proceeding from 98 % to 90 %. In
contrast, the isothermal dehumidifying process from B to C experienced a major change of the relative
humidity varying from 90 % to 40 %. From the point of view of the dehumidifying period, the cooling
dehumidification and isothermal dehumidification experienced 44 min and 145 min, respectively.
Therefore, the isothermal dehumidifying process is dominated during the operation of the TED.
The above experimental characteristics of the dehumidifying process depended on condensing heat
transfer on cold-side fin heat exchangers of the TED. Figure 4 shows the photos of cold-side fins before
and after dehumidification. It can be found that there existed many condensed liquid drops on the
extended surfaces. At the bottom of local fins, there appeared the liquid-bridge phenomenon. Such an
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
462
adherence of condensed liquid drops on the fin surfaces has an important role in the dehumidifying
process of the TED, which can be further explained as follows.
-5
0
5
10
15
20
25
30
35
40
0 20 40 60 80 100 120 140 160 180 200
Time
(
min
)
Temperaure (
o
C )
30
40
50
60
70
80
90
100
Relative humidity (%
)
Cabinet temperature
Cold fin temperature
Heat fin temperature
Relative humidity
Figure 2. A typical dehumidifying process of the TED
Figure 3. The dehumidifying process in the enthalpy-humidity diagram
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
463
(a) before dehumidification (b) after dehumidification
Figure 4. Photos of cold-side fins before and after dehumidification
At the initial dehumidifying process, the temperature of cold-side fins dropped rapidly, which promoted
the vapor condensation inside the cabinet. This indicated that a number of liquid drops was created and
adhered on the fin surfaces. Soon these liquid drops became a thin layer of liquid film, driven by the
surface tension. Such a liquid film was a great thermal resistance and then led to a weak heat and mass
transfer between the fin surfaces and the ambient inside the cabinet. As a result, the dehumidifying rate
began to decrease, and this was the so-called first decelerating stage ranging from 90% to 70%. When the
film became thicker and thicker, the runoff flow along the gravity direction was dominated. The fin
surfaces became relatively clean again. At the same time, the temperature of cold-side fins still remained
a steady low temperature close to 0
o
C. Thus, the dehumidifying process began to accelerate for the
second time. Repeatedly, when the liquid film on the fin surfaces became thick enough, the
dehumidifying process experienced the second decelerating stage till the dehumidification was ended.
From the analysis above, it can be seen that a rapid elimination of condensed liquid-drops on the cooling
fins amounted on the thermoelectric module is a major approach to improving the performance of the
TED. Such efforts can be realized by using superhydrophilic (SH) or super-water repellent (SWR)
surfaces [11-12]. Related experimental work is expected to be reported in our later studies.
3.2 The analysis of the dehumidifying performance of the TED
Figure 5 compares the dehumidifying processes under different input powers of the thermoelectric
module. It can be seen the dehumidifying period depends strongly on the input power of the
thermoelectric module. In this experiment, when the input power was 6.0 W, the dehumidifying period
reached the minimum of 120 min. In contrast, for the input powers lower or higher than 6.0 W, the
dehumidifying period showed an increasing tendency. For instance, the dehumidifying period was 180
min at 5.0W and 150 min at 7.0W, respectively. This result can be explained by the characteristics of
thermoelectric module materials.
For a thermoelectric module, the heat absorption from the ambient consists of two sources. One is from
the Peltier effect, which is proportional to the current through the module. Another is from the Joule
effect, which is proportional to the square of the current through the module [4]. When the input power is
too low, the Peltier effect is weak, resulting in a low dehumidifying rate. On the contrary, when the input
power is too high, the Joule effect is dominated, resulting in a high heat load imposed on the heat-cold
heat exchangers. Once the heat generated can not be released into the ambient, the extra heat will be
transferred by heat conduction towards the cold side, and finally weaken the condensing heat transfer on
the fin heat exchanger. Therefore, for the minimum dehumidifying period, there is an optimal input
power of the thermoelectric module.
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
464
0 20 40 60 80 100 120 140 160 180
30
40
50
60
70
80
90
100
110
Relative humidity (%)
Time (min)
4.0W
5.0W
5.5W
6.0W
7.0W
8.0W
Figure 5. Dehumidifying processes under different input powers of the thermoelectric module
Figure 6 compares the variations of COP under different input powers of the thermoelectric module. It
can be seen that the COP of the TED increases rapidly during the initial stage, then keeps a slowly
growing tendency till reaching the maximum level, and finally drops quickly. For instance, when the
input power was 6.0 W, the COP began to accelerate at 80 min, which was the start of the second
accelerating stage shown in Figure 5, and reached the maximum of 0.32 at about 100 min, which was just
the end of the second accelerating stage shown in Figure 5. For the most dehumidifying period, however,
the COP remained an average level of about 0.17, about half of the maximum level. When the input
power increased to 7.0 W and 8.0 W, the corresponding COP reduced to 0.26 and 0.20, respectively.
This experimental phenomenon can fully reflect the working mechanism of the TED where the complex
heat and mass transfer related with condensation on the cold-side fin heat exchanger is a critical
influencing factor. Compared with the previous results obtained by Chen [7] and Sumrit [8], the present
experimental COP was higher.
0 20 40 60 80 100 120 140 160 180
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
COP
Time (min)
4.0W
5.0W
6.0W
7.0W
8.0W
Figure 6. Variations of COP under different input powers of the thermoelectric module
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
465
45678
0.12
0.14
0.16
0.18
0.20
2
3
4
5
6
Average COP
Average dehumidifying rate
Average dehumidifying rate
(10
-3
g/min)
Average COP
Input power (W)
Figure 7. Variations of the average COP and the average dehumidifying rate under different input powers
of the thermoelectric module
Figure 7 further shows the variations of the average COP and the average dehumidifying rate under
different input powers of the thermoelectric module. In the present experiment conditions, the average
dehumidifying rate was about 0.005 g/min at the input power of 6.0W, though the maximum reached
0.0097g/min. It should be noted that the performance of practical TEDs varies greatly in the present
studies. A generally acceptable COP of TEDs ranges from 0.1 to 0.6[13]. In addition, Vian [14]
established a numerical model of a TED. Through optimizing the heat transfer, a COP of 0.8 was
obtained in the computation conditions. As far as the author’s knowledge, this value is the highest in the
available literatures. Therefore, compared with Vian’s theoretical prediction, there is still a great space
for the performance improvement of the present TED. In contrast, the COP of conventional
dehumidifiers usually ranges from 0.7 to 1.3 [15]. On the other hand, at present the figure of merit (ZT)
of most commercial thermoelectric materials is lower than 1.0. For instance, the ZT of BiTe used in this
experiment was about 0.6. However, the value of ZT has been improved to 2.4 for some novel
thermoelectric materials [16]. It can be expected that, with a continuing advance in the thermoelectric
technology, the performance of the TED will be improved drastically. In summary, despite that the COP
obtained by the present experimental setup is at a relatively low level, a lower power TED is of certain
commercial interest. From the present experimental results, it can be seen that, during the use of
thermoelectric cooling technology, the relative humidity can be controlled more accurately through an
adjustable input power, which provides an important approach to improving the indoor thermal comfort
of buildings.
4. Conclusion
TEDs have a promising potential in the situations where both a lower power and a smaller space are
required. The present work was performed to develop a low power TED as an alternative to the
conventional vapor-compression refrigeration systems. From the experimental results discussed above,
the following conclusions can be obtained:
(i)
The TED experienced two typical processes including the cooling dehumidification and the
isothermal dehumidification, but the latter was dominated. The relative humidity experienced two
accelerating stages and two decelerating stages. This phenomenon is related with the condensing
behavior on the extended surfaces of the cold-side fin heat exchanger.
(ii)
There exists a peak during the variation of the COP as a function of the input power of the
thermoelectric module. Under the present experimental conditions, the COP of the TED reached
the maximum of 0.32 and the corresponding dehumidifying rate was 0.0097 g/min, when the input
power was kept at 6.0 W.
(iii)
There is still a great space for the performance improvement of the present TED. The effective
elimination of condensed liquid-drops on the cold-side fins is a major work in the future.
International Journal of Energy and Environment (IJEE), Volume 1, Issue 3, 2010, pp.459-466
ISSN 2076-2895 (Print), ISSN 2076-2909 (Online) ©2010 International Energy & Environment Foundation. All rights reserved.
466
Acknowledgements
This work was supported by Program for Young Talents in Hebei University of Technology, and Natural
Science Foundation of Tianjin (No. 07ZCKFSF00400).
References
[1] Bansal PK, Martin A. Comparative study of vapour compression,thermoelectric and absorption
refrigerators. International Journal of Energy Research, 2000, 24(2), 93-107.
[2]
Alhazmy MM. Power estimation for air cooling and humidification using exergy analysis.
International Journal of Exergy, 2006, 3(4), 391-401.
[3]
Riffat SB, Qiu G. Comparative investigation of thermoelectric air-conditioners versus vapour
compression and absorption air-conditioners. Applied Thermal Engineering, 2004, 24(14-15),
1979-1993.
[4]
Rowe DM. Handbook of Thermoelectrics. Boca Raton, CRC Press, 1995.
[5]
Yang R, Chen G, Kumar AR, et al. Transient cooling of thermoelectric coolers and its applications
for microdevices. Energy Conversion and Management, 2005, 46(9-10), 1407-1421.
[6]
Riffat SB, Ma Xiaoli. Thermoelectrics: a review of present and potential applications. Applied
Thermal Engineering. 2003, 23(8), 913-935
[7]
Chen S.Condensation analysis and improvement of a thermoelectric dehumidifier.M.A.Sc. Thesis,
National Cheng Kung University, 2001.
[8]
Sumrit I. Development of thermoelectric dehumidifier. M.A.Sc. Thesis, King Monkue University
of Technology, 2003.
[9]
Bejan A, Tsatusaronis G, Moran M. Thermal design and optimization, Wiley, 1996.
[10]
Cengel YA. Heat transfer- a practical approach, McGraw-Hill, 1998.
[11]
Takata Y. Photo-induced hydrophilicity and its applications to phase change phenomena. Thermal
Science and Engineering. 2002, 10(6), 31-38.
[12]
Takata Y, Tanaka K, Kaijima K, et al. Enhancement of heat transfer with liquid-vapor phase
change by photo-induced hydrophilicity. Proceedings of the 33rd National Heat Transfer
Conference, Japan, 1999.
[13]
Huang BJ, Chi CJ, Duang CL. A design of thermoelectric cooler. International Journal of
Refrigeration.2000, 23(4), 208-218.
[14]
Vian JG, Astrain D, Dominguez M. Numercial modelling and a design of a thermoelectric
dehumidifier. Applied Thermal Engineering. 2002, 22(4), 407-422.
[15]
Pietro M, Francesco M, Daniele P. HVAC dehumidification systems for thermal comfort: a critical
review. Applied Thermal Engineering, 2005, 25(5), 677-707.
[16]
Rama V, Edward S, Thomas C, et al. Thin-film thermoelectric devices with high room-
temperature figures of merit. Nature, 2001, 413(6856), 597-602.
H. Wang is Ph.D in Thermal Engineering from Tianjin University in 2008. He completed M.Sc and B.Sc
in Engineering Thermophysics from Tianjin University in 2003 and 1998, respectively. At present, he is an
associate professor of School of Energy and Environment, Hebei University of Technology. His wor
k
focus mainly on heat and mass transfer related with renewable energies, HV&AC, and energy saving o
f
buildings. He has published over 20 research papers including 8 in the international journals. He takes as a
reviewer of several international journals in the energy field.
E-mail address: huajunwangl@126.com
C. Qi is Ph.D in Thermal Engineering from Tianjin University in 1997. He completed M.Sc and B.Sc in
Engineering Thermophysics from Tianjin University in 1988 and 1985, respectively. He is a professor o
f
School of Energy and Environment, Hebei University of Technology. Since from 2005, he has taken as the
dean of School of Energy and Environment of HBUT. His work focus mainly on heat and mass transfe
r
related with renewable energies, HV&AC, and energy saving of buildings. He has published several
research papers in the international journals.
E-mail address: qicy@mail.hebut.edu.cn