Thermoelectric properties and efficiency measurements under large temperature

differences
A. Muto, D. Kraemer, Q. Hao, Z. F. Ren, and G. Chen

Citation: Review of Scientific Instruments 80, 093901 (2009); doi: 10.1063/1.3212668
View online: http://dx.doi.org/10.1063/1.3212668
View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/80/9?ver=pdfcov
Published by the AIP Publishing

Articles you may be interested in
New experimental methodology, setup and LabView program for accurate absolute thermoelectric power and
electrical resistivity measurements between 25 and 1600 K: Application to pure copper, platinum, tungsten, and
nickel at very high temperatures
Rev. Sci. Instrum. 85, 095121 (2014); 10.1063/1.4896046

An apparatus for concurrent measurement of thermoelectric material parameters
Rev. Sci. Instrum. 84, 013907 (2013); 10.1063/1.4789311

Performance characterization of high-efficiency segmented Bi2Te3/CoSb3 unicouples for thermoelectric
generators
AIP Conf. Proc. 1449, 467 (2012); 10.1063/1.4731597

Thermoelectric Properties of Bi0.85Sb0.15 Alloys Prepared by High Press Method
AIP Conf. Proc. 824, 43 (2006); 10.1063/1.2192332

Progress in the development of high efficiency segmented thermoelectric generators
AIP Conf. Proc. 458, 1403 (1999); 10.1063/1.57535

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP:
158.42.28.33 On: Thu, 11 Dec 2014 09:52:09

and k共T兲 are the Seebeck coefficient. tions of the TE legs and radiation losses from the leg side sured independently over small temperature differences. and both meth- ods can result in large uncertainties in ZT̄共T兲.8.” The three temperature- dependent properties or “intrinsic properties” ␣共T兲. respectively. and such usages can potentially increase the overall of a TE element can be expressed as3. often greater Recent years have witnessed a trend of rising electricity than 10%.33 On: Thu.9 Often. However. published online 1 September 2009兲 The maximum efficiency of a thermoelectric generator is determined by the material’s dimensionless figure of merit ZT. considered such as electrical contact resistance at the junc- man method2 and 共2兲 the three intrinsic properties are mea. accepted 4 August 2009. temperature.3 walls. F. © 2009 American Institute of Physics.13. as well as the conversion efficiency. Chestnut Hill. The realistic testing conditions of this technique make it ideal for material characterization prior to implementation in a real thermoelectric generator.1 Q.5.2. Such testing conditions provide a more ond method uses different measurement systems for each realistic examination of actual TE device performance and individual property. and thermal conductivity of a single element or leg. The directly measured power and efficiency are compared to the values calculated from the measured properties and agree within 0. Boston College.11 efficiency in many systems. must be used to obtain the actual device efficiency. INTRODUCTION The second method is very time consuming. Chen1 1 Department of Mechanical Engineering. Muto.aip. power generators.28.1 The performance of a TE material is characterized most TH − TC 冑 1 + ZT̄ − 1 ␩max = 共1兲 commonly by its so-called dimensionless figure of merit TH 冑 1 + ZT̄ + TC/TH .14 The original Harman method is used to measure ␳共T兲 and In this work we develop methods to characterize a single ZT̄共T兲 over small temperature differences. measured efficiency.3212668兴 I. We has many variations and has been applied to bulk modules aim to extract the three intrinsic properties of a single leg and and thin films alike. and T̄ is the mean abso. respectively. ZT̄ = ␣2T̄ / ␳k or “intrinsic ZT. and where TH is the hot side temperature and TC is the cold side thermal conductivity. USA 共Received 26 June 2009. We measure the Seebeck coefficient. the maximum efficiency ␩max heat. USA 2 Department of Physics. strongly with temperature along the length of the element or ture dependent. Cambridge. Thermoelectric TE power generators usually operate under large tem- 共TE兲 devices can be used either as heat pumps for localized perature differences up to hundreds of degrees centigrade. electrical resistivity. REVIEW OF SCIENTIFIC INSTRUMENTS 80. Massachusetts 02467.1063/1. Reuse of AIP content is subject to the terms at: http://scitationnew. Ren. all three properties are not should thus support the development and design of future TE measured on the same sample or along the same direction. research progress at the device level must be made to validate materials and to guide system design. since the intrinsic properties vary lute temperature.4 This technique TE leg operating under a large temperature difference. The intrinsic properties are highly tempera. Real thermoelectric material properties are highly temperature dependent and are often measured individually using multiple measurement tools on different samples. Kraemer.7 The sec. 093901-1 © 2009 American Institute of Physics This article is copyrighted as indicated in the article. In this work we present an experimental technique that eliminates some of these uncertainties.4% and 2%. under a large temperature difference of 2 – 160 ° C.org/termsconditions.12 Further- Typically one of two methods is used to characterize a more.2 and G. Hao. Massachusetts Institute of Technology.00 80. and to measure each represents a formidable “leg. ␳共T兲. Downloaded to IP: 158. Massachusetts 02139. 093901 共2009兲 Thermoelectric properties and efficiency measurements under large temperature differences A. 共1兲 is not applicable and often numerical solutions task. As a result.1 Z.” Eq.10 costs and an emphasis on energy efficiency.6 The drawbacks are that it only works verify the results by comparing the predicted power conver- over small temperature differences and requires adiabatic sion efficiency from the property measurements to the actual boundary conditions that can be difficult to satisfy. in real TE applications additional effects must also be TE material over a small temperature difference: 共1兲 the Har. 0034-6748/2009/80共9兲/093901/7/$25. 11 Dec 2014 09:52:09 . If environmental control or heat engines for harvesting waste ZT̄ is temperature independent. electrical conductivity. 关doi:10. The advantages of this technique include 共1兲 the thermoelectric leg is mounted only once and all measurements are in the same direction and 共2兲 the measured properties are corroborated by efficiency measurements. reported ZT values have large uncertainties.3. For TE power generators to be successful.1 D.42.

OC and QC. or can be calculated by integrating the intrinsic prop.OC − Qrad.28. 093901-2 Muto et al. which radiation terms and can be solved to give expressions for includes QH. TC兲 and k共T兲 shown that since the temperature difference across the leg is large it is useful to characterize the TE leg in terms of its effective properties. Seebeck voltage. A QC. Reuse of AIP content is subject to the terms at: http://scitationnew. Qrad is the additional heat transfer rate on the hot side erties over the temperature interval. and the ra- diation loss.OC共TH. 共5兲 trical current I enters from the top and exits through the bottom of the sample. We will Equation 共4兲 is reduced to only the heat conduction and the directly measure the energy balance over the sample.3 The resistance mea- dx surement is applied in the limit that the current is small such that the Joule heat term can be neglected and the modulation 1 dVTE ␣ dT J=− − .side.OC are measured by a cali- The constitutive relations for one-dimensional heat flux brated heater on the hot side and a heat flux sensor on the q and current density J are as follows:15 cold side.aip. which will be described in more detail in Sec. TC兲. ␳共T兲. TC兲 and ment of a TE leg yields ␳共T兲 by This article is copyrighted as indicated in the article. L TC • “Effective properties” ␣eff共TH .TC兲⌬T = 冕 TC TH ␣共T兲dT. the open circuit heat transfer rates QH. and ⌬T ranges from 2 ° C to 160 ° C. QC. During the Figure 1 is a general schematic of a p-type TE sample properties measurement I = 0. and finally solve for the intrinsic properties. Illustration of a p-type TE leg under working conditions.OC共TH. Sci.side ⬍ 2% of QH.org/termsconditions. P is the pe- VTE TE rimeter. tion. 冕 • “Device properties” are the directly measurable values. it follows from Eq. 共2兲 method where VTE is the sense voltage. The leg is kept Material under high vacuum 共5 ⫻ 10−5 Torr兲 such that natural convec- tion and air conduction can be neglected. and − Qrad. 80. Governing equation of a Seebeck voltage and an Ohmic voltage drop. QC. + x=L TC B.” QH. TC兲. Instrum.42. The geom- the materials at each temperature as previously defined. Measuring these device properties directly as a function of temperature II. The heat transfer rates QH.side = kdT + Qrad trical resistance. tion. ZT̄eff共TH . ␴sb is the Stephan–Boltzmann constant. dT The electrical resistance is measured by a four-wire ac q = JT␣ − k . 共3兲 frequency is high such that the Peltier heat alternates sign ␳ dx ␳ dx and cancels due to periodic heating and cooling at the junc- An energy balance over a one-dimensional differential ele. 共7兲 ␳eff共TH . It will be keff共TH . Rev. 11 Dec 2014 09:52:09 . in.33 On: Thu. The elec. TC兲 and ␣共T兲 by from the top at TH and the cold side heat transfer rate QC exits the bottom at TC. VS共TH. TH cluding the heat transfer rates.side is the total • “Intrinsic properties” ␣共T兲. FIG. and VTE is the TE voltage measured across the leg at the position of the thermocouples. In general VTE is composed A. Here A is the cross sectional area. Here. is related to ␳eff共TH . and the electric power Pe = IVTE. Downloaded to IP: 158. resulting in an “effective ZT.side .OC and Qrad ⬍ Qrad.TC兲 = QH.OC. and k共T兲 are properties of radiation loss from the side walls 共see Appendix兲. and the electrical resistance of the leg R. We define three classes of interrelated proper- 共6兲 ties. Device properties measurement under large ⌬T QC I The measured device properties are the open circuit See- beck voltage VS. conduction. and T⬁ is the ambient temperature.TC兲 = ␣eff共TH.TC兲⌬T = A L 冕 TC TH kdT + Qrad .TC兲 = A L keff共TH. the Joule heat. etry of the leg was chosen to minimize the effects of radia- which will be extracted from the effective properties. MATHEMATICAL FORMULATION enables us to determine the effective properties. 共3兲 that under test. The electrical resistance R. ␧ is the emis- sivity. III. The temperature is measured by ther- mocouples. keff共TH . due to radiation from the side walls and Qrad. where the hot side heat transfer rate QH enters VTE = VS and is related to ␣eff共TH . and elec. A 共4兲 TH x=0 with four terms on the left hand side representing the heat . 093901 共2009兲 QH I A 冉 冊 d dT k dx dx − IT d␣ 2 ␳ dx + I − ␧␴sbP共T4 − T⬁4 兲 = 0. 1. where Qrad. the Thompson heat. TC兲.OC. TC兲 are obtained directly from device properties measurement when the hot side is at TH and cold side at TC.

another method would be to distinction is made between the intrinsic properties of the measure the power dissipated over an adjustable resistor.28. the TE leg. must be measured when electrical current flows through the leg. based on the effective properties. R and ␤ are the temperature integrated quantities ciency measurement to control TC independent of T⬁ in order described above and QH.6 mm3. Reuse of AIP content is subject to the terms at: http://scitationnew.33 On: Thu. no to control I and measure VTE.aip. If we approximate ␣共TH兲 ⬇ ␣共TC兲 = ␣eff. for calibration. 1 to solve for the temperature gradient. Thus Unlike the device properties measurement which is con- ZT̄eff is a useful concept for the design engineer because it is ducted under the open circuit condition. 093901-3 Muto et al.16. and the 1 1 QH = ITH␣共TH兲 + QH. Instrum. device-ready leg with metallization on both ends. and its purpose is to allow soldering Power is produced when the voltage has increased in the with low electrical and thermal contact resistances while act- direction of current flow. The cold assembly is soldered to the ␤共TH. 2共b兲兴 and calibrated for heat loss. Downloaded to IP: 158. Sci.6⫻ 1. 共12兲 surrounding vacuum chamber. TC兲 whereby we have neglected the radiation term to approxi.17 bulk and the properties of the contact. an electrode. The hot assembly electrode is very done in Eq. If significant contact Both methods are equally valid.42. ␣eff 2 T̄ V2ST̄/兩⌬T兩 ZT̄eff = = .TC兲 = ␳eff共TH. the Joule heating IVHeater will be balanced by the radiation fined by the intrinsic thermal conductivity alone as we have and conduction losses. Most samples that were mea- 共4兲 twice with respect to x and apply the boundary conditions sured had electrical contact resistances estimated to be ⬍5% illustrated in Fig.18 but remember that in mounted to a TE cooler module which was used in the effi- this case. Effective ZT ␳dT L 兰TC k␳dT TC TH L 1 R共TH. 093901 共2009兲 冕 D. In this way we use a current source ing as a diffusion barrier. If ␳. One require- where the electrical power output Pe can be obtained from ment for testing a TE sample is that the sample must be a Eq.11 we can define ZT̄eff共TH . The hot assembly is soldered 2 2 to the top of the leg and consists of a calibrated electrical where the Thompson heat is written as heater providing QH. The sample was a Bi2Te3 p-type leg temperature gradient is then used to evaluate QH and QC from Marlow Industries with dimensions of approximately from Eq. Prior to mounting the leg. the metallization material most commonly used is Ni.org/termsconditions. 1 but a few complications exist. 共8兲. In the same way ZT̄共T兲 was de- 共8兲 rived for intrinsic properties3. the efficiency ␩ a measure of the TE material quality over a large ⌬T.13 In the following analysis. 共10兲 Bi2Te3-based TE elements.OC − I2R − I␤ . they will be automatically lumped rent source method has many advantages.side + Pe Conceptually the experimental system is similar to that shown in Fig.11. Td␣ / dT are independent perature. a thermocouple measuring TH. a calibrated ther- mopile type heat flux sensor measuring QC. then the Thomp- A A TH dT A 兰TTHkdT C son heat term drops out and Eq. 共11兲 the device properties and energy conversion efficiency of the 2 2 TE leg. For Pe = IVS − I2R = IVTE .OC. The electrodes sidewall radiation.TC兲 = ⬇ . which is constant. 1. mate the temperature gradient as dT / dx ⬇ −Q / kA. 11 Dec 2014 09:52:09 . 共9兲 can be written in terms dx of the effective properties. thin to minimize heat conduction losses but this means it This article is copyrighted as indicated in the article. Efficiency ciency 关Eq. 共13兲 cold side of the leg and consists of a copper heat spreader. the cold assembly. 80. The entire assembly is contained in a vacuum chamber throughout the material such that half goes to the hot side at a pressure of 5 ⫻ 10−5 Torr and a uniform ambient tem- and half goes to the cold side.OC + I2R + I␤ . The of the total resistance. 共2兲. of temperature then the factors of 21 are exact 共see Appendix兲. resistances are present. T⬁ of 22– 24 ° C. and an electrode conducting I. We integrate Eq. The factors of 21 in front of the Joule heat supply current to the TE leg for the resistance and efficiency and Thompson heat terms are the result of an approximation measurements and are controlled by an external power sup- that assumes both heating terms are distributed uniformly ply. into the effective properties. The system is composed of the following elements: the hot assembly.6⫻ 1. In this case second approximation that the temperature gradient is de. The efficiency is defined as Pe Pe III. 共1兲兴 and maximum coefficient of performance of a real device by using the conventional ZT formulas. a thermocouple measuring TC. QC.OC include the effects of to reduce heat losses at the cold assembly. Rev. 1 1 Figure 2共a兲 illustrates the system designed to measure QC = ITC␣共TC兲 + QC. 共14兲 ␳effkeff C RQH It can be used to solve for the maximum conversion effi- C. the hot assembly was sus- When solving for R during power generation we make a pended 关Fig. The bottom side of the heat flux sensor is erty heat transfer rate equations3. 共9兲 QH QC + Qrad. and a heater used Equations 共11兲 and 共12兲 appear similar to the constant prop. 共3兲.TC兲 = 冕TC TH T d␣ dT dT. EXPERIMENTAL SYSTEM ␩= = . where Q = 共A / L兲兰TTHkdT is the conducted heat. but experimentally the cur.

TC兲 = y共T兲dT = f共TH兲 − f共TC兲. 3. The voltage response of the flux sensor was Based on the integral relationship between the intrinsic measured for different heat fluxes imposed by the calibration properties and the device properties.28. 093901-4 Muto et al. and then measure VS. 24 Then the heater power is reset and the properties are mea. during the bly are negligibly small. 3兲. Y共TH . Reuse of AIP content is subject to the terms at: http://scitationnew. − y共TC兲 . 80. The obtained emissivity of ␧ = 0. in order to capture the entire tem- duction along the electrode. 34 32 IV. We take the derivative with respect to Joule heating in QC but not in QH. produces a relatively large unknown Joule heating term for complexity. QC. We found that the the electrode was measured by mounting another heater to latter method was more accurate. Taking the centered solved for the emissivity. Radiation losses at the cold assem. dTH dTH ⳵ TC TH dTH sembly heat conduction loss was dominant to Joule heating dTC in the electrode. 11 Dec 2014 09:52:09 .6 40 agrees reasonably with Fourier transform infrared spectros- Cold Side Temperature TC (°C) 38 copy measurements from similar samples and in addition is evidence that the heater and flux sensor have been accurately 36 calibrated. differences where QC.org/termsconditions. In our calibration it was found that heat losses from the dY共TH.OC.33 On: Thu. TC兲 and an intrinsic property y共T兲 related by 冕 trode is negligible compared with the heat conduction along TH the electrode. Due to temperature drifts in the system.OC ⬍ 0. Although it is Hot Side Temperature TH (°C) possible to control the cooler module for each successive heater power setting to keep TC constant. which is much thicker than the perature range from ⌬T of 2 ° C to 160 ° C. Instrum. QH.aip. Sci. which is mounted to the side of the cold assembly.1 W. 共Color online兲 A schematic diagram of the 共a兲 property and efficiency measurement system and 共b兲 calibration procedure for the hot side heater and the cold side heat flux sensor. and R from Sec.OC hot assembly electrode.side = QH. 冏 冏 QC during the efficiency measurement.OC for small temperature efficiency measurement.OC with ⌬T = 75 ° C and and evaluate the derivatives numerically. The electrical resistance between the positions Y共TH. wait for the system to reach steady 26 state. the intrinsic properties heater.TC兲 df共TH兲 ⳵ f共TC兲 dTC = − = y共TH兲 hot assembly constituted about 10% of QH. and the combined correction value consti. 093901 共2009兲 Cryostat Chamber at T Cryostat Chamber at T a) b) Hot Assembly IVHeater IVHeater Hot Assembly TH TH VTE I TE IVHeater TE TE TC TC HEATER HEATER HEATER Cu Spreader I IVHeater Cu Spreader I I Cold Assembly Flux Sensor Cold Assembly Flux Sensor Cu Spreader II Cu Spreader II TE Cooler VFlux TE Cooler VFlux Cold Finger Cold Finger FIG. This article is copyrighted as indicated in the article. PROPERTY DATA ANALYSIS 30 The procedure for the properties measurement is to set a 28 constant heater power. therefore we rely only on T H.42. this increases FIG. 共16兲 tuted ⬍3% of QC. We correct for where ⳵ f共T兲 / ⳵T = y共T兲. which allows us to solve for the TC Joule heating inside the electrode when I ⫽ 0.OC I ⫽ 0 and thus QH cannot be accurately measured during the had larger uncertainties than QH. 2. Rev. Therefore. At the cold as. Cold side temperature vs hot side temperature during the device steady state settling times considerably and adds unnecessary properties measurement. we use QH. II B.OC − QC. side circuit. 共Fig. We performed an energy balance over the dTH leg such that Qrad. Downloaded to IP: 158. The primary heat loss is from con. in which the radiation loss of the elec. 共15兲 of TE and TC was measured.OC and allow TC to rise with increasing TH the cold assembly electrode was disconnected from the out. 20 40 60 80 100 120 140 160 180 200 sured again for different values of TH and TC. can be solved for by regression fitting or taking the numeri- After the flux sensor calibration. The process can be under- the end of the electrode and measuring the temperatures TE stood by considering an arbitrarily measured device property and TC 关Fig. properties measurement. the thermal conductance of cal derivative of the device properties. In order to calibrate the flux sensor instead of QC. 2共b兲兴.

Rev.42. Downloaded to IP: 158.8 ZT 16 0. 3 erties following the procedures described in Sec.6 14 0. IV.TC) Seebeck Coefficient (V/K) 240 2. and 0. given TH where T̄H = 共TH. 11 Dec 2014 09:52:09 . 共c兲 electrical resistivity. To verify that the intrinsic property curves accurately in the low temperature range because they will have been differentiated correctly we used the intrinsic prop- also affect the higher temperature range. For instance.aip.TC) 180 (T) 1. T (°C) FIG.4%. 093901 共2009兲 a) b) 250 2. 80. difference. k共T兲. Reuse of AIP content is subject to the terms at: http://scitationnew. This article is copyrighted as indicated in the article.6 eff(TH. 1 calculated Eq.9%.4 k(T) 230 2.TC 1. 5.2 + TH. and ZT̄共T兲 as well as their calculated Eq.08 so it is important to evaluate the properties ZT̄ = 0.TC) 12 ZT(T) 10 0.6 Thermal Conductivity (W/mK) keff(TH. II was evalu- ated by taking the numerical derivative of the device prop. 4. Sci. 6 Maximum Efficiency (%) 5 V. 共17兲 ⌬TH ⌬TH sign engineer to quickly determine the performance of a TE generator made with this leg. 共b兲 thermal conductivity.8 200 190 1. 3 that ⌬TC / ⌬TH ⬇ 0. PROPERTY RESULTS 4 Each of the intrinsic properties from Sec. and 共d兲 dimensionless TE figure of merit. We observe from Fig.2 20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160 180 200 Temperature TH.0 20 (T) Resistivity m 18 0. we solve for the arbitrary property as a function ZT̄ drops off with increasing temperature more rapidly than of temperature ZT̄eff due to the integral relationship between the intrinsic ⌬Y ⌬TC and effective properties.28. T (°C) c) d) 22 effTH. T (°C) Temperature TH. (1) measured sured effective properties are plotted as discrete points with error bars while the intrinsic properties have been plotted as 0 0 20 40 60 80 100 120 140 160 180 linearly interpolated curves.2 20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160 180 200 Temperature TH.5 ° C.4 170 1.4 ZTeff(TH. Instrum. 共Color online兲 Maximum efficiency vs temperature difference across 0.2 220 2.1兲 / 2 is the mean temperature of two = 198 ° C and TC = 39 ° C 共Fig.0 210 1. The four TE properties vs temperature and their respective effective properties vs hot side temperature: 共a兲 Seebeck coefficient.org/termsconditions. The uncertainty was estimated Temperature Difference T (°C) individually for each effective property point. The maximum uncertainty values for ␣eff. and ␳eff are approximately FIG. T (°C) Temperature TH. respectively. 3兲. keff. A linear regression fit was applied to the first few measurement points in the lower temperature range. Figure 4共d兲 shows that the leg for a cold side temperature of 24.7%. 0.64 material. ␳共T兲. the leg should operate as a consecutive measurements.33 On: Thu. Figures 2 4共a兲–4共d兲 plot ␣共T兲. The mea. (9) respective effective properties as a function of TH. 093901-5 Muto et al. This graph can be used by the de- y共T̄H兲 = + y共T̄C兲 .

org/termsconditions. Integrating Eq.4 0.28. the Thompson heat should be included in the efficiency procedure was to apply a constant heater power and current.7 6. the estimated uncertainty was less than large temperature difference ranging from 2 ° C to 160 ° C.0 5. to calculated efficiency in Fig. The property mea. side wall radiation. + C1 = 0. 共9兲兴 versus current at both temperature The authors acknowledge financial support from MAS- ranges. 6. with a temperature ranges.0 5. Sci.5 0. we are able to write approximate solutions The goal of the efficiency measurement is to firmly es. the points with error bars are the measured values. is not well satisfied.3 1.42.4 1. 80.0 1. DOE 共Grant No. Using effec- VI.5 5.5 30 3.4% FG02-08ER46516兲 for the property measurement.9 6. calculation for most accurate results. The tures. to model a single TE leg for power generation. The small version efficiency was 2.9 1. for the power and efficiency. Instrum.5 5. Rev. the lines are calculated from the measured properties using Eqs. 共Color online兲 Efficiency and power vs current.0% while that for Fig. the fact that the measured power and effi- ciency are in excellent agreement with each other is convinc. but it is still a significant excellent agreement for ⌬T ⬍ 120 ° C. 0. 6共b兲 is 0. Reuse of AIP content is subject to the terms at: http://scitationnew. The measurements include the effects of intrinsic property data using Eq. with significantly less uncertainty TC = 24. 093901 共2009兲 a) b) 3.8%–5.2 1. dT I2 kA + ␳dx⬘ − I ␶dT⬘ − ␧␴sbP 共T4 − T⬁4 兲dx⬘ surement uses QH. The larger relative difference of efficiency in Fig. Downloaded to IP: 158. Therefore.8 0.3%.1 26 2. and for all measured points. 共9兲 and 共10兲. DE- agreement for both cases with a relative difference of ⬍0. 共1兲 with the calculated ZT̄eff from Eq.OC and the efficiency measurement uses dx A 0 TH 0 QC. erties to solve for ZT̄eff at the same measured temperature VII.5 ° C. The power and efficiency divergence at high ⌬T reflects that ␣共T兲 drops rapidly at high were directly measured and are within 0. CONCLUSION points as in Fig. contact resistance.aip. The technique measures all three TE properties of a single Figure 5 plots the maximum efficiency versus ⌬T for leg. in the same direction. This where ␶ = Td␣ / dT is the Thompson coefficient and C1 is a independent check feature is an important advantage of the constant of integration. Equations 共1兲 and 共9兲 show of the hot side heat transfer rate. 6共a兲 is likely an indication of the larger APPENDIX uncertainty in QC at small heat transfers rates.3 28 2.3 7.5%.OC and QC are independently measured x T x quantities with independent calibrations. The efficiency was measured at two different rectly measuring an energy balance over a single leg.5 0. Side wall radiation loss constituted ⬍2% crete points are measured values. Pe.33 On: Thu. 共b兲 Hot side temperature is 169– 172 ° C. If the See- leg and to compare these results to the calculated power and beck coefficient varies strongly over the operating tempera- efficiency from the intrinsic properties measurement. 4共d兲 and found the relative difference to be In this work.1%–0. EFFICIENCY RESULTS tive properties. 共14兲兴. which confirms that it part of the energy balance considering that the power con- is appropriate to characterize a material by ZT̄eff. measure the steady state QC. Figures 6共a兲 and 6共b兲 plot the measured and calculated ACKNOWLEDGMENTS efficiency 关Eq. and Eq. The relative difference of measured KFUPM for modeling. 6共a兲 is 0. We apply a one-dimensional model considering temperature-dependent material properties and side wall radiation loss. which are mathematically simi- tablish what performance can be expected from the particular lar to the well-known constant property formulas. 共4兲 once with respect to x gives 冕 冕 冕 Recall that QH. and then repeat for another We developed an experimental technique capable of di- current setting. The solid curve is the calculated value from the than other methods. the dotted curve is using temperature dependent properties.7 Current (A) Current (A) FIG. 11 Dec 2014 09:52:09 .6 0. cold side is 24– 25 ° C.4% and 2% of the temperature and the assumption ␣共TH兲 ⬇ ␣共TC兲 = ␣eff.1 7. we employ an effective property approach 0. values calculated based on the property measurements. 093901-6 Muto et al. 关Eq.2 27 2.5 5.6 1. 共9兲. 共14兲 and the dis.6%. We integrate a second time with re- described technique. 共a兲 Hot side temperature is 86– 93 ° C. spect to x and apply the boundary conditions T 兩x=0 = TH and This article is copyrighted as indicated in the article.2%.5%–2.7 0.4 29 Efficiency (%) Efficiency (%) Power (mW) Power (mW) 5.5 1. 共A1兲 ing evidence that both have been measured accurately. The calculated and measured powers are in excellent DAR for the efficiency measurement.

New York. that the quantities in the parenthesis are approximately equal 1995兲. 共2兲 to derive Eq. 34. and B. J. 1986兲. Caillat. M. Boca Raton. 282–285.” U. 11 D. Nature L兰L0 ␳dx 冉 冊 共London兲 413. Snyder. J. N. and M. to 21 . Gao and D. 共A4兲 ference on Thermoelectrics (ICT 2003).33 On: Thu. 1261 共2001兲. 1976兲. Reuse of AIP content is subject to the terms at: http://scitationnew. Phys. p. Jr. M.. 17 P. 2003兲. Rep. Phys. W.28. J. Sci. 共12兲. Zoltan. dT A TH ␧␴sbP L x 4 H. Electronic Refrigeration 共Pion. Proceedings of the 18th International Conference on Ther- heat conduction term and a radiation term given by moelectrics. J. Iwasaki and H. 2005兲. J. Ure. Part 1 41. H. D. D. Ohkawa. J. Iwasaki. O’Quinn. W. Harman. 13. D: Appl. J. Koyano. 冏 冏 冕 冕冕 3 H. J. Qrad = dT⬘dT 共IEEE. 6606 − kA = kdT + 共T4 dx TH L TC L 0 0 共2002兲. then we can apply this result to Eq. 10 where ␶ = Td␣ / dT is the Thompson coefficient. pp. L TC T dT⬘ dT 15 N. Department of Energy. Diller and L. J. London. M. 11 Dec 2014 09:52:09 . 共A5兲 − T⬁4 兲dx⬘dx − I2 LA 冕冕 0 L x 0 ␳dx⬘dx Both radiation quantities have been approximated by assum- ing a linear temperature gradient. 1 共1958兲. If we assume M. Sherman. Phys. 931 共2004兲. P. and A. − I␤ 共A3兲 H H . 7 兰TTC 兰TT⬘ ␶dT⬘dT H. Siivola. 80. P. 1 共1960兲. This article is copyrighted as indicated in the article. Phys. pp. Hori. 1351 ent at the hot side 共1959兲. d. Fleurial. New York. T. 8. Rowe. Appl. Instrum. D. Phys. E. v. G. Rev. 093901 共2009兲 T 兩x=L = TC to solve for the integration constants Qrad. Pauw. 共A2兲 1 T. Proceedings ␧␴sbP TH TH 共T⬘4 − T⬁4 兲 of the 20th International Conference on Thermoelectrics (ICT 2001). Appl. R. Tanaka. J. Department of Electrical Engineering and Computer Science This is the additional hot side heat transfer at x = 0 due to side 共MIT.org/termsconditions. Thermoelectrics. and H. Hendricks.S. W. Appl. L兰TTC ␶dT 9 H J. Sci. the leg to solve for the total side wall radiation loss 1301 共2004兲. New York. Satake. and K. 13 T. Terasaki. FL. Goldsmid. J. Cahn. Zoltan. + I L 冕冕 TC TH T TH ␶dT⬘dT. 31. G.. Appl. E. Proceedings of the 22nd International Con- ␧␴sbPL TH TH ⬇ 共T⬘4 − T⬁4 兲dT⬘dT. 8 L. 冉 冊 5 A. 269. and R. and A. Hori. wall radiation. Proceedings of the 24th International Conference on Thermoelectrics (ICT 2005) 共IEEE. 473–476. 2001兲. and I. Applying I = 0 results in Eq. Phys. “Engineering Scoping Study of Thermoelectric Generator Systems for Industrial Waste Heat Recovery. Bell. Appl. M. M. Ashroft and N. 8. which is composed of a Borshchevsky. ⌬T2 TC T 571–573. 1999兲. Colpitts. 冕 冕 14 T. Downloaded to IP: 158. We integrate the radiation along the length of 18 M. New York. Borshchevsky. Rowe. Cape and G. Mayer. Kontostavlakis. Meas. R. A. Lehman. Logan. 597 共2001兲. C. Solid State Physics 共Brooks/Cole Th- dx dx omson Learning. 共6兲. T.aip. Mermin.42. H. Jpn. pp. pp. 1909 共1963兲. 513–516.side = ␧␴sbP 冕 TC 共T4 − T⬁4 兲 dT ⬇ ␧␴sbPL 冕 TH 共T4 dT I2 − kA = dx A 冕 0 x ␳dx⬘ − I 冕 T TH ␶dT⬘ − ␧␴sbP 冕0 x 共T4 TH dT dx ⌬T TC − T⬁4 兲dx⬘ − A L 冕 TC TH kdT + ␧␴sbP L 冕冕 0 L x 0 共T4 − T⬁4 兲dT. J. Fleurial. J. Caillat. Phys. Cambridge. A. Venkatasubramanian. 冕 冕 16 R. Rowe. 30. 2 T. 093901-7 Muto et al. 兰L0 兰x0␳dx⬘dx 96. − T⬁4 兲dx⬘dx − I2R 6 R. Fujii. W. 共IEEE. Snyder. Heikes. Gao. CRC Handbook 共CRC. Philips Res. After rearranging terms we solve for the temperature gradi. 2007兲. 2006. 12 B. Technol. 共IEEE. T.