Accepted Manuscript

Experimental study of the effects of the thermal contact resistance on the per-
formance of thermoelectric generator

Shixue Wang, Tianxi Xie, Hongxi Xie

PII: S1359-4311(17)32982-4
DOI: https://doi.org/10.1016/j.applthermaleng.2017.11.036
Reference: ATE 11406

To appear in: Applied Thermal Engineering

Received Date: 2 May 2017

Revised Date: 10 October 2017
Accepted Date: 5 November 2017

Please cite this article as: S. Wang, T. Xie, H. Xie, Experimental study of the effects of the thermal contact resistance
on the performance of thermoelectric generator, Applied Thermal Engineering (2017), doi: https://doi.org/10.1016/
j.applthermaleng.2017.11.036

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 Experimental study of the effects of the thermal contact

resistance on the performance of thermoelectric generator

Shixue Wanga,b Tianxi Xiea Hongxi Xiec

a
School of Mechanical Engineering, Tianjin University, Tianjin 300072, China
b
Key Laboratory of Efficient Utilization of Low and Medium Grade Energy, Tianjin University, Tianjin

300072, China
c
Hefei General Machinery Research Institute, Anhui 230031, China

Abstract: Semiconductor thermoelectric generation technology is promising for application to waste

heat recovery and is increasingly attracting research interest. The objective of the present study was to

determine how the performance of a thermoelectric generator (TEG) was affected by the thermal

contact resistance. The effects of the interface material and loading pressure—factors that determine

the thermal contact resistance—on the TEG performance were experimentally investigated and

analyzed. It was found that, for a given temperature difference between the cold and hot sources, the

actual temperature difference of the thermoelectric module and the output power significantly increased

when an interface material was applied to the contact surface. In addition, with increasing loading

pressure, there was a decrease in the thermal contact resistance between the two sides of the

thermoelectric module, and an increase in both the actual temperature difference across the module and

the output power. Further, for a given loading pressure, an increase in the thermal resistance of the

thermoelectric module was accompanied by an increase in the average temperature of the module, a

decrease in the ratio of the thermal contact resistance to the total thermal resistance, and an increase in

the output power.

Keywords: thermoelectric generator; thermoelectric module; thermal contact resistance; loading

pressure; interface material

*
Corresponding author

Tel.: +86-22-27402567; fax: +86-22-27402567; email: wangshixue_64@tju.edu.cn.

1
NOMENCLATURE

A cross-sectional area ( m2) TL temperature of the cold side of the

Q heat flow through the contact interface thermoelectric module (K)

Q1 heat flow through the hot source (W). ZT dimensionless figure of merit (–)

Q2 heat flow through the cold source (W) T temperature difference (K)

QA average heat flow through the hot and U open voltage (V)

cold side (W) t/ temperature gradient (K/m)

P output power (W) Greek letters

PMAX maximum output power (W)  Seebeck coefficient (V/K)

pl loading pressure (Pa)  thickness (m)

r electrical resistance  efficiency

R total thermal resistance (K/W)  heat conductivity (W/(Km))

RC thermal contact resistance (K/W)  ratio of the thermal contact resistance to

the total thermal resistance
Subscripts
Re Reynolds number

RM thermal resistance of the thermoelectric a average

module (K/W) C contact

RCM sum of the module thermal resistance and H hot side of the thermoelectric module
thermal contact resistance l loading
RCM = RC + RM (K/W) L cold side of the thermoelectric module

T temperature (K) M module

T1 temperature of the hot source (K) MAX maximum

T2 temperature of the cold source (K) 1 hot source

TH temperature of the hot side of the 2 cold source

thermoelectric module (K)

1. Introduction

Transportation vehicles powered by internal combustion engines are one of the main causes of

energy crises and environmental pollution. From an energy balance perspective, approximately

30%–45% of the fuel burned in such vehicles is released into the air as exhaust [1]. The development of

a means of recycling and reusing the heat energy contained in the exhaust would thus contribute to

improving energy efficiency and reducing environmental pollution. Talom et al. [2] briefly introduced

2
some heat recovery technologies applicable to vehicle engines. The technologies use a thermoelectric

generator (TEG), which is an environmentally friendly energy conversion device, to directly convert

thermal energy into electrical energy. The many merits of a TEG include compactness, high reliability,

and zero emission

The heat recovery efficiency of TEG technology in current practical applications is relatively low.

To enhance the economic viability, it is necessary to improve the thermoelectric conversion efficiency.

This requires the development of new high ZT thermoelectric materials [3–5] and optimization of the

thermoelectric module structure [6–7]. Moreover, the improvement of heat transfer on the hot and cold

side heat exchangers of a TEG enhanced the effective temperature difference between both sides of the

thermoelectric module [8–12]. Tzeng et al. [8] developed a semiconductor thermoelectric power

generation system that used a metal pin-fin array as the heat absorber and heat sink. The system was

found to afford significantly higher TEG performance. Lu et al. [9] investigated the effects of metal

foam on the performance of a TEG that is used for automobile waste heat recovery. They found that,

under given conditions, low-porosity and low-pore-density metal foams significantly improved the

output power and thermoelectric conversion efficiency. Lesage et al. [10] examined the effects of the

channel wall geometric structures on the heat transfer characteristics of a TEG. The results of their

experiments showed that, compared with spiral inserts, inserts with protruding panels afforded more

effective improvement of the thermoelectric power production. Wang et al. [11] used a symmetrical

staggered arrangement of dimpled surfaces on the upper and lower surfaces of the heat exchanger of a

TEG. They achieved enhanced heat transfer at high Re values, resulting in increased output power of

the TEG. Zhang et al. [12] investigated the performance of a TEG that utilized phase change heat

transfer. The system included an asymmetric heat transfer surface between the exhaust and the hot side

of the thermoelectric module, which was filled with a gas-liquid phase change material. The

evaporation and condensation of the phase change material was used to realize efficient heat transfer,

and optimization of the area ratio enabled the achievement of a significantly enhanced the output

power.

In particular, stable thermal contact between the TEG and the hot source, and between the TEG

and the cold source significantly impact the system performance. Considering that no surface is

perfectly smooth, an air gap exists between the two interfaces. In addition, the large difference between

the thermal conductivities of air and solids causes contraction of the heat flow through the contact

3
interface, resulting in the generation of a heat transfer resistance, i.e., the thermal contact resistance.

The existence of the thermal contact resistance decreases the actual temperature difference between the

two sides of the thermoelectric module, thereby reducing the output power. Previous studies [13–14]

have shown that the thermal contact resistance may be decreased by using thermal grease to eliminate

the air gaps at the contact interfaces. This is because of the higher thermal conductivity of the grease

compared to that of air. Goodarzi et al. [15] investigated and quantified the effect of a nano-aluminum

coating of the contact interface on the thermal contact resistance, and found that it significantly

enhanced the heat transfer performance by decreasing the thermal contact resistance by 38%. Further,

for the given hot and cold temperatures and thermal interface material, the loading pressure has been

identified as another major determinant of the thermal contact resistance [16].

Several other studies have been conducted on different aspects of a TEG, including reduction of

the thermal contact resistances between the TEG and the hot and cold sources, improving the overall

performance of the system, and establishing the formation mechanism of the thermal contact resistance

and its effect on the performance of the TEG system. For example, Rowe et al. [17] investigated the

mechanism and influencing factors of the thermal contact resistance of a TEG, and found that the

thermal contact resistance significantly restricted the system performance. Astrain et al. [18] examined

the performance effects of the thermal contact resistances between a TEG and its hot and cold sources.

They found that a 10% reduction of the thermal contact resistances between the two sides of the heat

exchanger increased the TEG performance by 8%. Sakamoto et al. [19] examined the effect of the

interface materials on the performance of a TEG system and determined the most suitable materials for

the hot and cold sides. They noted that the use of the interface materials reduced the thermal contact

resistance between the thermal interfaces, thereby increasing the performance of the thermoelectric

generation system. Du et al. [20] investigated the effect of the loading pressure on the TEG

performance and found that an increase in the loading pressure decreased the thermal contact resistance

of the thermoelectric module and increased the temperature difference across the element. Whereas

they established an overall increase in the performance of the TEG, the change in the thermal contact

resistance was not quantified.

Most of the previous investigations of the effects of the thermal contact resistance on the

performance of a TEG were qualitative. The present study was thus aimed at a quantitative practical

assessment of the effect of the thermal contact resistance of a TEG system. The determinant factors of

4
the thermal contact resistance were also investigated, as well as the mechanisms of theirs effects on the

TEG performance. In addition, a method for reducing the thermal contact resistance was presented.

Based on the findings of the study, this paper proposes a design basis and guidance for improving the

performance of a TEG system.

2. Effect mechanism of the thermal contact resistance on a TEG

The present study was based on the following basic assumptions: (1) The Seebeck coefficient and

electric resistance of the thermoelectric module do not vary with the temperature and loading pressure.

(2) The thermal resistance of the thermoelectric module does not vary with the loading pressure. (3)

The thermal contact resistance between the interfaces does not vary with the temperature.

Further, it was generally assumed that the temperature of the TEG system was continuous at the

interfaces between the thermoelectric module and the hot and cold sources. In reality, because the

contact surfaces were not flat, temperature difference occurred at the contact interfaces. The thermal

contact resistance at a contact interface can be modeled by the one-dimensional steady heat conduction

equation:

RC=ΔTC/Q (1)

where RC is the thermal contact resistance at the interface (K/W), ΔTC is the temperature difference

between the thermoelectric module and either the hot or cold source (K), and Q is the heat flow through

the contact interface (W).

The heat flow can be calculated by the one-dimensional steady heat conduction equation for a

solid material in contact with the hot side of the thermoelectric module:

t
Q1  A (2)

where  is the thermal conductivity of the solid material (W/(Km), A is the cross-sectional area (m2),

and t/ is the temperature gradient of the solid material (K/m).

The open circuit voltage U of a TEG can be calculated by the following formula:

U = (TH-TL) (3)

where  is the Seebeck coefficient of the thermoelectric module (V/K), and TH and TL are the

temperatures of the hot and cold sides of the thermoelectric module (K), respectively.

The maximum output power PMAX of a TEG can be calculated by the following formula:

PMAX=U2/4r (4)

5
where r is the electrical resistance of the thermoelectric module ().

The presence of the thermal contact resistance at the contact interfaces and the resultant increase

in the thermal resistance to the heat transfer process are accompanied by a decrease in the effective

temperature difference on both sides of the thermoelectric module. If the changes in the Seebeck

coefficient and the electrical resistance of the thermoelectric material are neglected, the open-circuit

voltage and maximum output power would only be affected by the effective temperature difference

between the two sides of the thermoelectric module. A large effective temperature difference would

enhance the TEG performance. It is assumed that, by neglecting the change in the thermal contact

resistance with temperature, the thermal contact resistances on both sides of the thermoelectric module

would be equal:

T1  TH TL  T2
 (5)
Q1 Q2

Q2  Q1  PMAX (6)

Here, T1 and T2 are the wall temperatures of the hot and cold sources, respectively (K), Q1 is the heat

flow from the hot source to the thermoelectric module (W), and Q2 is the heat flow from the

thermoelectric module to the cold source (W).

By experimental measurement of U, Q1, T1, and T2, the actual hot-side temperature TH and cold

side temperature TL of the thermoelectric module can be determined by equations (3)–(6). Using these

parameters, the thermal contact resistances between the TEG and the hot and cold sources, as well as

the thermal resistance of the thermoelectric module, can be determined.

The sum of the thermal resistance and the thermal contact resistances of the thermoelectric module

can be calculated using

T1  T2 （7）
RCM 
Qa

where RCM is the sum of the thermal resistance and the thermal contact resistances (K/W), and Qa is the

average heat flow through the hot and cold sides of the thermoelectric module (W).

3. Test apparatus

Figure 1 shows a schematic of the experimental setup, which consists of (1) a voltage regulator, (2)

the test system, and (3) a data acquisition system. The test system consists of, from bottom to top, a

heater, trapezoidal copper block, rectangular stainless-steel block, thermoelectric module, heat sink,
6
pressure sensor, and loading pressure devices.

Figure 1
The voltage of the heater was adjusted by a voltage regulator to vary the heating power, and hence

the temperature of the hot side of the thermoelectric module. The trapezoidal copper block was used to

increase the heat flow through to its large heat absorption area and small heat release area. The high

thermal conductivity of copper enhanced the uniformity of the heat source temperature field. To

improve the measurement accuracy of the temperature distribution, a stainless steel block (56 × 56 × 35

mm) with a relatively low thermal conductivity was used in the device for measuring the heat flow of

the TEG heat source. Five K-type thermocouples with an outer diameter of 0.5 mm were inserted

horizontally into the stainless-steel block along its centerline, axially spaced at 5 mm. The

thermocouple closest to the heated surface was 2 mm from it. The five thermocouples were calibrated

using a platinum resistance thermometer. The wall temperature and heat flow were calculated by

applying Fourier’s law, with the thermal conductivity of the stainless steel evaluated for the arithmetic

average of the temperature measurements. The linearity of the temperature distribution was confirmed.

The employed K-type thermocouples had an accuracy of ±0.1 ℃.

The thermoelectric module (TEHP-24156-1.2, Hi-Z Technology) used in this study consisted of

126 Bi2Te3 based HZ-20 thermoelectric modules connected in series, with overall dimensions of 56 ×

56 × 5 mm. The heat sink was a rectangular channel measuring 100 × 100 × 20 mm. The 20 °C cooling

water flowing through the heat sink constituted the cold source of the TEG. The cold-source

temperature was measured by a K-type thermocouple inserted into a hole on the top of the heat sink

close to the cold side of the thermoelectric module. The hole was of diameter 0.5 mm and depth 50 mm.

Among other benefits, the water cooling reduced the heat transfer resistance of the heat sink, improved

the heat flow, increased the temperature difference between the two sides of the thermoelectric module,

and enhanced the measurement precision. The loading pressure was measured by a pressure sensor

(MIK-LCLY) with a measurement range of 0–2 MPa and full scale accuracy of 0.03%.

To reduce the heat loss, the outer surfaces of the heater and the trapezoidal copper and stainless

steel blocks were covered with insulation material. During the experiments, the temperatures of both

sides of the thermoelectric module were controlled by regulating the power of the heater and the flow

rate of the cooling fluid with respect to the loading pressure. The corresponding temperature,

7
open-circuit voltage and other data were acquired. All the data were recorded and stored in the

GL900-8-UM-851 data acquisition system.

4. Results and discussion

To determine the effect of the thermal contact resistance on the performance of the TEG, the

performance was measured and analyzed under steady state conditions for various loading pressures

and heat flow, with and without applying thermal grease to the contact interface

4.1. Effect of the interface material

Figure 2
Figure 2 shows the variations of the temperature difference between the two sides of the

thermoelectric module with the temperature difference between the cold and hot sources of the TEG for

a loading pressure of 326 kPa with and without applying thermal grease to the contact interface. As we

can be observed from the figure, the temperature difference across the thermoelectric module increases

with increasing temperature difference between the cold and hot sources. Moreover, for a given

temperature difference between the cold and hot sources, the temperature difference across the

thermoelectric module is significantly higher when thermal grease is applied to the contact interface.

Application of thermal grease to the contact interface eliminates the air gap and thus decreases the

thermal contact resistance, resulting in a higher ratio of the thermal resistance of the thermoelectric

module to the total thermal resistance.

Figure 3
Figure 3 shows the variations of RCM with the heat source temperature for a loading pressure of

326 kPa with and without the thermal grease at the interface. It should be noted that, after the water

cooling, the temperatures of the cold side of the thermoelectric module and the cold source undergo

very little changes. As can be observed from Figure 3, RCM increases with increasing heat source

temperature. Considering the assumed negligible variation of the thermal contact resistance with

temperature, for a given loading pressure, the variation of the RCM with temperature is the same as that

of the thermal resistance of the thermoelectric module. The specific trends of the latter are shown in

Figure 7. In addition, the RCM when using the thermal grease at the interface is significantly less than

that without the thermal grease. This is because the thermal grease eliminates air from the gap between

the contact surfaces, thereby decreasing the thermal contact resistance.

8
 Figure 4
Figure 4 shows the relationships between the output current and the output power and

thermoelectric conversion efficiency, respectively, for loading pressure of 326 kPa and 148 K

temperature difference between the hot and cold sources. As can be seen from the figure, regardless of

whether thermal grease is applied to the contact interface, the output power and thermoelectric

conversion efficiency vary parabolically with increasing output current. However, the use of the

thermal grease significantly increases the output power and conversion efficiency. Specifically, it

increases the maximum output power from 1.44 W to 2.32 W, and the conversion efficiency from 1.09%

to 1.23%, representing enhancements of 61% and 12.8%, respectively.

4.2. Effect of the loading pressure

Figure 5
Figure 5 shows the variation of RCM with the loading pressure at different heat flow with applying

grease to the contact interface. As can be observed from the figure, RCM initially decreases dramatically

with increasing loading pressure, especially between 7.6 and 108 kPa. Because of the presence of the

air gap at the contact interface, the thermal contact resistance is relatively large. However, with

increasing loading pressure, the gaps at the contact interfaces between the TEG and the hot and cold

sources decrease rapidly, the effective contact area at the contact interface increases, and the thermal

contact resistance also decreases rapidly. With further increase of the loading pressure, there is a drop

in the rate of decrease of RCM. When the loading pressure exceeds 548 kPa, RCM becomes almost

constant with further increase of the loading pressure. This is because the gap at the contact interface

becomes very small, with further increase of the loading pressure having negligible effect on it, hence

the almost constant RCM.

In this study, the variation in the thermal resistance of the thermoelectric module with the loading

pressure was neglected; hence, the variation of RCM with the loading pressure was the same as that of

the thermal contact resistance. The actual temperatures on the two sides of the thermoelectric module

can be determined using equations (3)–(6), while the thermal contact resistance can be determined by

equation (1). The variation of the thermal contact resistance with the loading pressure is shown in

Figure 6.

Figure 6
9
 Figure 6 reveals that the thermal contact resistance initially decreases drastically with increasing

loading pressure, with the rate of decrease gradually decreasing until the thermal contact resistance

becomes constant. During the increase of the loading pressure from 7.6 to 765 kPa, the thermal contact

resistance decreases from 0.423 to 0.257 K/W, as can be observed from Figure 5. It is also noteworthy

that, for a given loading pressure, the thermal contact resistance is almost independent of the heat flow

(the hot side temperature of the module or average module temperature). This implies that it is

reasonable to neglect the variation of the thermal resistance of the thermoelectric module with the

loading pressure, as well as the variation of the thermal contact resistance with temperature, as was

done in this study.

Figure 7
Figure 7 shows the variation of the thermal resistance of the thermoelectric module with the

average temperature of the module. As can be observed, the thermal resistance increases with

increasing average temperature. The experimental data reveal a smooth variation, with only a little

deviation. The results indicate that it is suitable for the thermal resistance of the thermoelectric module

not to vary with the loading pressure. The loading pressure only affects the average temperature of the

thermoelectric module. The observed variation of the thermal resistance of the thermoelectric module

with the heat flow within the measurement range of the present experiments is in good agreement with

the indications of the thermoelectric module manufacturer [21]. This confirms the accuracy of the

experimental measurements.

Figure 8
Figure 8 shows the relationships between the output current and the output power and conversion

efficiency for a temperature difference between the hot and cold sources of 148 K and various loading

pressures. As can be observed, the output power and thermoelectric conversion efficiency both vary

parabolically with the output current. In addition, the output power reaches maximum when the load

resistance is equal to the resistance of the thermoelectric module, and the output power and

thermoelectric conversion efficiency increase with increasing loading pressure. For a loading pressure

of 765 kPa, the maximum output power is 2.46 W, which corresponds to the maximum conversion

efficiency of 1.32%. When the loading pressure drops to 109 kPa, the output power drops to 1.85 W,

which corresponds to the conversion efficiency of 1.1 %. Hence, the maximum output power and

10
conversion efficiency are respectively improved by 33% and 20% when the loading pressure is

increased from 109 to 765 kPa. This is because, for a given temperature difference between the hot and

cold sources, an increasing loading pressure decreases the thermal contact resistance, resulting in an

increase in the actual temperature difference between the two sides of the thermoelectric module, and

hence enhancements of the output power and thermoelectric conversion efficiency.

4.3. Comparison of the effect of increasing heat transfer

Figure 9

In an actual transportation vehicle with a TEG exhaust heat recovery system, the presence of

thermal resistance in the heat transfer processes significantly affects the system performance. Figure 9

shows the thermal resistance network of a practical TEG system. The thermal resistance between the

hot air and the environment comprises the following: (1) convection thermal resistance between the hot

fluid and the air-side heat exchanger (Rh1), (2) thermal resistance to the heat conduction through the

air-side heat exchanger (Rw1), (3) thermal contact resistance between the heat exchanger and the TEG

(Rc1), (4) conduction thermal resistance of the thermoelectric module (RM), (5) thermal contact

resistance between the TEG and the heat sink (Rc2), (6) thermal conduction resistance of the heat sink

(Rw2), and (7) thermal convection resistance between the heat sink and the cooling fluid (Rh2). Because

of the existence of thermal resistance for each heat transfer process, the thermal resistance

decreases during any heat transfer process will enhance the TEG performance.

To investigate the effect of increasing the heat transfer, the heat transfer conditions of the hot and

cold sources in references [8] and [12] were applied to the thermoelectric module of the present study.

The determined heat transfer coefficient and thermal resistance at the hot end of the module during the

heat transfer process are presented in Table 1.

Table 1

Figure 10
Figure 10 shows the relationship of the total thermal resistance at the hot side of the thermoelectric

module and the ratio of the thermal contact resistance on the hot side to the total thermal resistance

with the loading pressure. The data were obtained based on the heat transfer conditions on the hot side

11
(neglecting the thermal resistance of the wall) in the works of Tzeng et al. [8] and Zhang et al. [12]. It

can be observed from the figure that the total thermal resistance decreases with increasing loading

pressure. Because the loading pressure only affects the thermal contact resistance during the heat

transfer process, the change in the total thermal resistance is equal to the change in the thermal contact

resistance. In addition, Figure 10 reveals that an increase in the heat transfer coefficient at the hot side

of the module is accompanied by an increase in the ratio of the contact resistance to the total thermal

resistance. In [12], the ratio of the contact resistance to the total thermal resistance is about 10%–12%,

whereas it is as high as 58% in [8]. Decreasing the thermal resistance to the heat transfer process can

thus be effectively used to improve the TEG performance, especially when a very high-performance

heat exchanger is used on the hot side.

5. Conclusions

The objective of this study was to determine the effect of the thermal contact resistance on a TEG

performance. For this purpose, the effects of the interface material, heat flow, and loading pressure on

the TEG performance were experimentally measured and analyzed. The experimental results showed

that the performance was enhanced by the use of an interface material, namely, thermal grease, and by

increasing the loading pressure. Following is a summary of the study findings:

1. The application of the thermal grease to the contact interface eliminated air from the interface

gap, thereby decreasing the thermal contact resistance and increasing the temperature difference

between the two sides of the thermoelectric module. This improved the performance of the TEG.

Specifically, for a temperature difference between the hot and cold sources of 148 K, the application of

the thermal grease significantly increased the maximum output power to 2.32 W, from 1.44 W when

the grease was not used. This represented an increase of the conversion efficiency from 1.09% to

1.23%.

2. With the application of the thermal grease, the contact resistances between the thermoelectric

module and the hot and cold sources decreased with increasing loading pressure, while the actual

temperature difference across the module increased. For a temperature difference between the hot and

cold sources of 148 K, the maximum power output and conversion efficiency of the TEG were

respectively enhanced by 33% and 20% when the loading pressure was increased from 109 to 765 kPa.

3. For a given loading pressure, the thermal resistance of the thermoelectric module was observed

to increase with increasing average temperature of the module, and the ratio of the thermal resistance of

12
the thermoelectric module to the total thermal resistance also increased. There were further increases of

the actual temperature difference between the two sides of the thermoelectric module, the maximum

output power, and the TEG conversion efficiency.

Acknowledgments

This study was financially supported by the International Cooperation Research Program of the

Ministry of Science and Technology of the People’s Republic of China (Grant No. 2016YFE0118600).

References

[1] C, Yu, K.T. Chau. Thermoelectric automotive waste heat energy recovery using maximum power

point tracking. Energy Conversion and Management, 50 (2009): 1506-1521.

[2] H.L. Talom, A. Beyene. Heat recovery from automotive engine. Applied Thermal Engineering.

29(2009): 439–444.

[3] A.I. Boukai, Y. Bunimovich, T.K. Jamil, et al. Silicon nanowires as efficient thermoelectric

materials. Nature, 451(2008): 168-171.

[4] G.S. Nolas, D.T. Morelli, T.M. Tritt, et al. A phonon-glass-electron crystal approach to advanced

thermoelectric energy conversion applications. Annual Review of Materials Science, 29 (1999):

89-116.

[5] T.C. Harman, M.P. Walsh. Nanostructured thermoelectric materials. Journal of Electrical

Materials, 34 (2005): 119-122.

[6] J.H. Meng, X.D. Wang, W.H. Chen. Performance investigation and design optimization of a

thermoelectric generator applied in automobile exhaust waste heat recovery. Energy Conversion

and Management, 120 (2016): 71–80.

[7] A. Haider, S.Y. Bekir, A. Abdullah. Innovative design of a thermoelectric generator with extended

and segmented pin configurations. Applied Energy, 187 (2017): 367–379.

[8] S.C. Tzeng, T.M. Jeng, Y.L. Lin. Parametric study of heat-transfer design on the thermoelectric

generator system. International Communications in Heat and Mass Transfer, 52 (2014): 97–105.

[9] C. Lu, S.X. Wang, C. Chen. Effects of heat enhancement for exhaust heat exchanger on the

performance of thermoelectric generator. Applied Thermal Engineering, 89 (2015): 270-279.

13
[10] F.J. Lesage, E.V. Sempels, N.L. Bertrand. A study on heat transfer enhancement using flow

channel inserts for thermoelectric power generation. Energy Conversion and Management, 75

(2013): 532–541.

[11] Y.P. Wang, S. Li, X. Yang. Numerical and Experimental Investigation for Heat Transfer

Enhancement by Dimpled Surface Heat Exchanger in Thermoelectric Generator. Journal of

Electrical Materials, 45 (2016): 1792-1802.

[12] X. Zhang, C. Lu, S.X. Wang. Numerical Simulation of Thermoelectric Generator With Phase

Change Exchanger. Journal of Engineering Thermophysics, 07(2015): 1515-1519.

[13] L.H. Zhan, X.Q. Li. Experiment Research for the interfacial Thermal Contact Resistance. Light

Alloy Fabrication Technology, 9 (2002): 40-43.

[14] C .Yuan, B. Duan, L. Li. Thermal conductivity and contact resistance measurements for thermal

interface materials. CIESC Journal, 66 (2015): 349-353.

[15] G. Koorosh, S.R. Reza, H. Shaaker. Reducing thermal contact resistance using nanocoating.

Applied Thermal Engineering, 70 (2014): 641-646.

[16] D.M. Bi, H.X. Chen. Influence of temperature and contact pressure on thermal contact resistance

at cryogenic temperatures. Cryogenics, 52 (2012): 403-409.

[17] D.M. Rowe. CRC Handbook of Thermoelectrics. New York：CRC Press, 1995.

[18] D. Astrain, J.G. Vian. Study of the influence of heat exchangers thermal resistances on a

thermoelectric generation system. Energy, 35(2010): 602–610.

[19] S. Tatsuya, I. Tsutomu. Selection and Evaluation of Thermal Interface Materials for Reduction of

the Thermal Contact Resistance of Thermoelectric Generators. Journal of Electrical Materials, 43

(2014): 3792-3800.

[20] Q. Du, Y.H. Zhang. Influence of Contact Pressure on the Performance of Thermoelectric

Generator. Journal of Tianjin University (Science and Technology), 47 (2014): 009-014.

[21] Specification of Bi2Te3-Based Thermoelectric Ingot. http://www.thermonamic.com.cn/Power_

Module nstallatio.pdf .

14
Figure captions

Fig. 1 Schematic diagram of the experimental system

Fig. 2 Variation of TM with T

Fig. 3 Variation of RCM with the heat source temperature

Fig. 4 Relationships between the output current and the output power and conversion efficiency

Fig. 5 Variation of RCM with the loading pressure

Fig. 6 Variation of the thermal contact resistance with the loading pressure

Fig. 7 Variation of the thermoelectric module thermal resistance with temperature

Fig. 8 Relationships between the output current and the output power and conversion efficiency

Fig. 9 Thermal resistance network of a practical TEG system

Fig. 10 Variations of the total thermal resistance and the ratio of the thermal contact resistance

with the loading pressure

15
Table captions

Table 1: Three different thermal boundary conditions

16
1 fixtures, 2 heater, 3 trapezoidal copper block, 4 stainless steel block, 5 thermoelectric module

6 heat sink, 7 pressure sensor, 8 Data acquisition, 9 PC, 10 voltage regulator, 11 power source

Fig.1 Schematic diagram of experimental system

17
Fig.2 Variation of TM with T

18
Fig. 3 Variation of RCM with heat source temperature

19
Fig. 4 Relationships between the output current and the output power and conversion efficiency

20
Fig. 5 Variation of RCM with loading pressure

21
Fig.6 Variation of contact thermal resistance with loading pressure

22
Fig. 7 Variation of thermoelectric module thermal resistance with temperature

23
Fig. 8 Relationship between output current and output power and conversion efficiency

24
Fig.9 Thermal resistance network of a practical TEG system

25
 5
Condensation
4 Staggered
In-line
3

R (K/W)
2

80
0

60
 (%)

40

20

0
0 100 200 300 400 500 600 700 800
pl (kPa)

Fig. 10 Variations of the total thermal resistance and the ratio of the thermal contact resistance

with the loading pressure

26
 Table 1: Three different thermal boundary conditions

Heat transfer Heat transfer Thermal resistance

Reference
method coefficient W/(m2K) (K/W)

Zhang et al [12] Condensation 2000 0.16

Fins（Staggered） 157.79 2.02

Tzeng et al [8]
Fins（In-line） 139.01 2.29

27
Highlights

The thermal contact resistance between the module and the hot and cold
sources were evaluated

The effects of the interface material and loading pressure on the

performance of TEG were exhibited

The ratio of the thermal resistance and the total thermal resistance in
different heat transfer means was revealed

28