development of a scalable 10% efficient thermoelectric generator

22
Development of a Scalable Development of a Scalable 10% Efficient Thermoelectric 10% Efficient Thermoelectric Generator Generator D. Crane, J. LaGrandeur, L. Bell DEER Conference, Detroit, MI August 15, 2007

Upload: trannguyet

Post on 14-Feb-2017

216 views

Category:

Documents


0 download

TRANSCRIPT

Development of a Scalable Development of a Scalable 10% Efficient Thermoelectric 10% Efficient Thermoelectric

GeneratorGenerator

D. Crane, J. LaGrandeur, L. Bell

DEER Conference, Detroit, MIAugust 15, 2007

Build on Recent SuccessesBuild on Recent SuccessesMaterials:

ZT substantially greater than unity has been demonstrated at places such as Michigan State, Research Triangle Institute (RTI), and MIT Lincoln Labs

Segmentation devices have successfully been demonstrated at Jet Propulsion Laboratory (JPL)

Thermodynamics and Heat Transfer:

Thermal isolation in the direction of flow that can improve cooling/heating performance by a factor of two

High power density designs that require 1/6 the thermoelectric (TE) material usage of conventional designs

TGM Development MethodologyTGM Development Methodology

Model, design and build fractional generator, first with low temperature TE material followed by high temperature TE material and finally segmented TE material

Validate performance model under varying operating conditions

Replicate fractional generator to scale up to >500 watts for low temperature TE material followed by similar scale up for high temperature (20W) and segmented TE materials

Integrate fractional generators with heat exchangers on the hot and cold side

Test and revalidate the performance model at varying operating conditions

Low Temperature BiLow Temperature Bi22TeTe33Subassembly Subassembly

Low temperature device built to continue learning process while high temperature material systems are being developedBi2Te3 thermoelectric generator module (TGM) output nominally targeted at > 500 watts

• Hot side fluid: oil, inlet temperature = 200 to 250C• Cold side fluid: water or glycol/water, inlet temperature = -5 to 30C

Peak Power = .969893

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0 2 4 6 8 10 12 14 16 18 20

Device Current (amps)

Elem

ent P

ower

(wat

ts)

n1p1n2p2n3p3

Test Results of Single ModuleTest Results of Single Module((TThh = 190C, water bath = 20C)= 190C, water bath = 20C)

Solid Model Isometric & Cross Solid Model Isometric & Cross Section ViewsSection Views

Cold manifold

Hot manifoldTE circuit between hot and cold manifolds

Expansion bellows

Thermoelectric SubassembliesThermoelectric Subassemblies

Assembled Single Hot Plate TE GeneratorAssembled Single Hot Plate TE Generator

Generator Test SetupGenerator Test Setup

Power Gen One Fifth Device

0

2

4

6

8

10

12

14

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0

Current (Amps)

Volta

ge (V

)

200C_-5C190C_-5C180C_-5C170C_-5C160C_-5C160C_6C160C_15C160C_25C131C_5C113C_5C98C_4C

27Jun07

Test ResultsTest ResultsMaximum Open Circuit Voltage >12VMaximum Open Circuit Voltage >12V

Power Gen One Fifth Device

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0

Current (Amps)

Pow

er (W

atts

)

200C_-5C190C_-5C180C_-5C170C_-5C160C_-5C160C_6C160C_15C160C_25C131C_5C113C_5C98C_4C

27Jun07

Test ResultsTest ResultsMaximum Peak Power of 130WMaximum Peak Power of 130W

Single plate(interfacial resistance = 2μΩcm2,

hot volume flow = 8 gpm (Xceltherm 600),cold volume flow = 8.85 lpm (water or glycol/water))

0102030405060708090

100110120130140150160170180

0 20 40 60 80 100 120 140 160 180 200 220 240 260

hot fluid inlet temperature (C)

pow

er (W

)

cold water inlet = 5C

cold water inlet = 15C

cold water inlet = 25C

cold glycol/water inlet = -5C

cold glycol/water inlet = 5C

cold glycol/water inlet = 15C

cold glycol/water inlet = 25C

cold water inlet = 4.68C, 8.85 lpm, hot oilflow = 4.3 gpmcold water inlet = 5.12C, 8.85 lpm, hot oilflow = 3.8gpmcold water inlet = 5.33C, 8.85 lpm, hot oilflow = 4.2 gpmcold water inlet = 6.60C, 8.85 lpm, hot oilflow = 7.5 gpmcold water inlet = 14.13C, 8.85 lpm, hot oilflow = 7.9 gpmcold water inlet = 24.51C, 8.85 lpm, hot oilflow = 8 gpmcold glycol/water inlet = -3.23C, 11.0 lpm,hot oil flow = 6.1 gpmcold glycol/water inlet = -6.02C, 10.9 lpm,hot oil flow = 6.0 gpmcold glycol/water inlet = -6.62C, 10.9 lpm,hot oil flow = 7.0 gpmcold glycol/water inlet = -5.68C, 10.9 lpm,hot oil flow = 6.9 gpmcold glycol/water inlet = -6.60C, 10.8 lpm,h t il fl 6 8

Peak Test Results Compared to SimulationsPeak Test Results Compared to SimulationsMeasured to simulated values vary by < 5%Measured to simulated values vary by < 5%

TE Figure of Merit for Current MaterialsTE Figure of Merit for Current Materials

Materials for segment layers are chosen to maximize ZT over each element’s exposed temperature range.

Materials of choice currently include p- and n-type skutterudite and Bi2Te3 as well as TAGS and n-type PbTe.

Ref: Modified from - http://www.its.caltech.edu/~jsnyder/thermoelectrics/

TE Couple Configuration TE Couple Configuration Alternatives with Segmented Alternatives with Segmented

ElementsElements

Traditional configuration

Alternative “Y” configuration

heat

heat heat

p-TAGS

p-CeFe3RuSb12

p-Bi2Te3

current

n-CoSb3

n-PbTe

n-Bi2Te3current

heat

heat heat

p-TAGS

p-CeFe3RuSb12

p-Bi2Te3

p-TAGS

p-CeFe3RuSb12

p-Bi2Te3

current

n-CoSb3

n-PbTe

n-Bi2Te3

n-CoSb3

n-PbTe

n-Bi2Te3current

p-CeFe3RuSb12

heat

heat

heat

p-TAGS

p-Bi2Te3

n-CoSb3

n-PbTe

n-Bi2Te3

current

p-CeFe3RuSb12

heat

heat

heat

p-TAGS

p-Bi2Te3

n-CoSb3

n-PbTe

n-Bi2Te3

current

heatheat

heatheat

heatheat

p-TAGS

p-Bi2Te3

n-CoSb3

n-PbTe

n-Bi2Te3

current

TGM Design ObjectivesTGM Design Objectives

State-of-the-art thermoelectric materialsThermally isolated elements in the direction of flowHigh power density designSegmented elements to maximize ZT over entire temperature rangeAbility to adjust segment thicknesses to increase TE material compatibility within elementsUse “Y” configuration to accommodate differing element thicknesses and differing thermal expansion coefficientsUse non-rigid joints to reduce the effects of thermal expansion mismatch

High Temperature TGM High Temperature TGM Developmental Prototype (20W)Developmental Prototype (20W)

TGM subassembly (cartridge heaters on hot side- full scale TGM will use hot gas heat exchangers) Solid model of fractional TGM

in its test fixture

20W Generator Build20W Generator Build

Completed halves of Completed halves of the 20W device, the 20W device, prior to the final prior to the final

assemblyassembly

Final assembled 20W Final assembled 20W device, plumbed and device, plumbed and

ready for testingready for testing

Test Results for the 20W Test Results for the 20W GeneratorGenerator

20W generator performance

(cold bath = 20C, six TAGS/PbTe couples, couple has 4 elements per side)(element dimensions = 3 x 3 x 2 mm)

(test #1 = solid, test #2 = dotted)

0123456789

1011121314151617181920212223

0 10 20 30 40 50 60 70 80 90 100

current (A)

pow

er (W

)

Th,avg = 149.0C, Tc,avg = 23.9CTh,avg = 149.6C, Tc,avg = 23.3CTh,avg = 198.2C, Tc,avg = 25.3CTh,avg = 200.3C, Tc,avg = 24.6CTh,avg = 247.8C, Tc,avg = 26.6CTh,avg = 249.5C, Tc,avg = 25.9CTh,avg = 297.2C, Tc,avg = 27.9CTh,avg = 299.3C, Tc,avg = 27.2CTh,avg = 347.6C, Tc,avg = 29.3CTh,avg = 348.8C, Tc,avg = 28.4CTh,avg = 398.5C, Tc,avg = 30.6CTh,avg = 398.1C, Tc,avg = 29.7CTh,avg = 446.6C, Tc,avg = 32.1CTh,avg = 448.8C, Tc,avg = 31.2CTh,avg = 471.8C, Tc,avg = 32.7CTh,avg = 474.7C, Tc,avg = 31.7C

Segmented TAGS/PbTeSegmented TAGS/PbTe--BiTe CoupleBiTe Couple

BiTe (0.6 mm)

PbTe (2mm)

BiTe (0.6 mm)

TAGS (2mm)

10% Efficiency Demonstration10% Efficiency DemonstrationPerformance is reproducible after Performance is reproducible after thermocyclingthermocycling

Segmented Couple PerformanceMarlow TAGS (Ti5Si3 metalization), PbTe, Bi2Te3

4 p-type, 6 n-type, TAGS (3 x 3 x 2mm), PbTe (3 x 3 x 2mm), Bi2Te3 (3 x 3 x 0.6mm)segments soldered together, alumina silica fiber paper insulation

01/10/07

0

1

2

3

4

5

6

7

8

9

10

11

0 100 200 300 400 500 600

temperature difference (C)

effic

ienc

y (%

)

cold bath = 20C (trial 1)

cold bath = -20C (trial 1)

cold bath = 20C (trial 2)

cold bath = 20C (trial 3)

cold bath = 20C (trial 4, unit turnedoff and restarted)

Next StepsNext StepsComplete assembly and testing of full-scale low temperature generator

Evaluate new design concepts that enhance performance and structural integrity of the device over the usage temperature range and during thermal cycling

Test fractional-scale prototype segmented element devices to analyze problem areas in the design, validate the model, and develop different segmentation techniques

Continue working with and testing new TE materials and new TE material combinations to push towards increasing average ZT

Continue working with partners to reduce interfacial resistance to 0.1 μΩcm2 at room temperature

Continue working on reducing heat losses from the system

Build full-scale high temperature TGM device using segmented elements

AcknowledgementsAcknowledgements

John Fairbanks and the DOE FCVT program

Aaron Yocum of DOE NETLAndreas Eder, Ryan Kottenstette, Boris

Mazar of BMWSteve Hung of VisteonJeff Sharp of Marlow Industries