U.S. Army Research, Development and Engineering Command

P. J. TaylorUS Army Research Laboratory

patrick.taylor7@us.army.milSensors and Electron Devices Directorate

301-394-1475ATTN: RDRL-SEE-I (Taylor)2800 Powder Mill RoadAdelphi, MD 20783

Thermoelectric Materials and Device Research at the Army Research Laboratory

Sept. 29, 2009

Outline

•

Army Interest and Applications•

External Program (RTI)

•

Internal Programs–

Cooling Devices

–

Remote Power Devices–

Thin-Film Thermoelectric Materials

–

Low-Parasitic TE measurments–

Seedling

•

Four Focus Areas -

Partnering•

Summary

Army Interests and Applications

••

ARL is the ArmyARL is the Army’’s R&D labs R&D lab

University OutreachFundamental ResearchApplied Research

ARL Directorates:C4ISRHuman Dimension Vehicle TechnologySensors & Electron DevicesSurvivability & Lethality AnalysisWMRDARO

•ARO•SEDD

Overlap -

DOE Freedom Car

•

Army Systems (usually) don’t:–

access cold sea water to reject heat–

have access to deep space to reject heat(sometimes can provide robust air flow)

–

offer choice of fuel (JP-8)

SEDD INTEREST:

•

TE Power Generation:–

Direct generation for soldier power (rchg batteries)–

Fuel utilization for vehicles–

Energy scavenging for small sensors

•

TE Cooling:–

Cooling of infrared detectors–

Temperature control for emitters

•

Army considers options:–

Solar cells–

Fuel cells–

Batteries-

detectability

External Program

“High-Performance Thermoelectric Energy Conversion and Energy Scavenging Demonstration”

••

GoalsGoals

Y1:Y1:

33--stage 50 Wstage 50 Wee

(15%)(15%)11--stage 25 Wstage 25 Wee

(collection)(collection)

Y2: 3Y2: 3--stage 250 Wstage 250 Wee

(15%)(15%)>exchanger integration>exchanger integration

11--stage 100 Wstage 100 WeeDesign 1kWattDesign 1kWattHeat Source

Heat Exchangerdesign 250 We

unit demo from fuel burner100 We

unit from exhaust or Army utilitiesPath to scale to 1 kilowatt for widespread use

••

Hard DeliverablesHard Deliverables

*

Internal Work

••

Goal: scalable (tandem), costGoal: scalable (tandem), cost--effective, covert, smalleffective, covert, small

••

Device Fabrication:Device Fabrication:––

PbTePbTe––

Burner Burner (Ivan Lee)(Ivan Lee)––

HeatHeat--sink sink (Brian Morgan)(Brian Morgan)

•

State-of-the-art performance–

>2 Watts/cm2

••

Issues:Issues:––

HJ electrical contactsHJ electrical contacts––

Better materialsBetter materials––

Thermal managementThermal management(heat(heat--sink)sink)

••

Small Devices for Remote PowerSmall Devices for Remote Power

5 mm5 mm

y = 9E-05x2 - 0.0027xR2 = 0.9954

0

1

2

0 50 100 150 200Temperature Difference (K)

Pow

er (W

atts

)

ARL PbTe Generatorvacuum (solid green) open-air (open)overnight (red)

Internal Work

••

Goal: MEMS integration (UGS)Goal: MEMS integration (UGS)––MBE/nanostructuringMBE/nanostructuring––LowLow--power levelspower levels––Infrared cloaking (?)Infrared cloaking (?)

••

SingleSingle--crystal IVcrystal IV--VI (PbTe) VI (PbTe) and Vand V--VI (BiVI (Bi22

TeTe33

) epilayers on silicon) epilayers on silicon

•

Shown:––Excellent crystallinity (Excellent crystallinity (Journ. Elec. MatJourn. Elec. Mat.).)––DopingDoping––Good TE propertiesGood TE properties––DryDry--etchingetching

••

Issues:Issues:––Thermal ExpansionThermal Expansion––Device Fabrication (metals)Device Fabrication (metals)––Contact resistanceContact resistance

••

ThinThin--Film TE Materials and DevicesFilm TE Materials and Devices(Morgan)(Morgan)

30 30 mm

9 9 mm

30 30 mm

9 9 mm

Dry-etched Bi2

Te3

array

B. MorganP.J. Taylor

TEM PbTe/PbSeTe NDSLunpublished: Moiré

fringeanalysis of nanodots, 2003

IV-VI

II-VI

Si

PbTe on 3”

Si

Internal Work

••

Thermoelectric MeasurementsThermoelectric Measurements

y = 247.3xR2 = 0.9987

0

500

10000 1 2 3

Temperature Difference (K)

Volta

ge (

V)

y = -178.93xR2 = 0.9905

-500

00 1 2

Temperature Difference (K)

Vol

tage

( V

)

Goal:Goal:

(Bi,Sb)(Bi,Sb)22

(Se,Te)(Se,Te)33

PbTePbTe--SimplifySimplify--parasiticsparasitics--device fab.device fab. Radiant Heat

(Seebeck)

Apply D.C.T0 @T=0 QPelt.

= Qrad. Qwire

& Qtc

=0(Know Qrad.

heat flow at T≠0)

Solve for (geometry)

I = 0.087 amperesSeebeck Coefficient ( meas )A/l = 0.405 cm2/cm

= 0.0149 W/cm-K

I = 0.057 amperesSeebeck Coefficient ( meas )A/l = 0.081 cm2/cm

= 0.0285 W/cm-K

MeasureResistivity @T=0 …No Seebeck error

BULK

Radiant Heat

I

Internal Work

••

SeedlingSeedlingPOC: Dr. Jeff Wolfenstine (301)-394-0317

Objectives:Develop nanostructured CoSb3

skutterudites by lithium ion transport

UNCLASSIFIED

Technical Challenge•Use electrochemical means to form amorphous/nanostructured n-type CoSb3

Desired Impact•Higher efficiency.

•Long-Term: improved energy conversion for

Army mobile power and energy recovery

Accomplishments

A

B

A: As-prepared: -Crystalline CoSb3B: After Electrochemical processing:

(one cycle discharge/charge) –Amorphous CoSb3

0

0.5

1

1.5

2

2.5

3

3.5

0 500 1000 1500 2000 2500

NSKD YbBa (C/12)June 16A (C/20)

Vol

tage

(V v

s. Li

)

Capacity (mAh/g)

A B B

w/ Prof. Jeff Sakamoto, Michigan State University

Li +

Li +

(Gross, Pazik, Rosker, DARPA, ARO)

Relevancy(devices)

Research Challenge 1:

TE Materials

Curves adapted from compilation by Tim Hogan (http://www.egr.msu.edu/%7Ehogant/Group%20Web%20Page/Research%20page.htm#thermoelectrics)

••

MaterialsMaterials

Research Challenge 2:

Thermal Management

Power (Power (ΔΔTTTETE

))22

The result of managingThe result of managingthe heat the heat FLOWFLOW

RRHX2HX2

RRHX1HX1

RRTETE

TotalHeat R

TQ

TTHOTHOT

TTCOLDCOLD

••

Heat Exchangers (HX) are critical:Heat Exchangers (HX) are critical:–

HX thermal resistance (RHX

) limits total heat flow (& Power)

–

Relative RHX

’s change temperature at thermoelectric device

–

Potentially dominates system size

••

Research paths:Research paths:–

System specific trade studies for Active vs Passive air and liquid cooling

–

Efficient heat spreading

(Morgan)(Morgan)

Research Challenge 3:

Packaging-Multistage

•

Many Thermal interfaces:–

Typically involve a grease–

Diminishing returns in cascaded devices targeting large ΔT & high efficiency

–

Grease or solders must survive high temperature

•

Scaling to large area / power:–

High thermal gradients create large stress as area increases

–

Huge CTE mismatches •

Bi2

Te3

~19 ppm/K•

SiGe ~4 ppm/K•

Thermal spreader ~4 ppm/K•

Heat Exchanger ~23 ppm/K–

Need improvements in package design & materials

Electrical insulator, thermal spreaderElectrical insulator, thermal spreader

Electrical insulator, thermal spreaderElectrical insulator, thermal spreader

Electrical insulator, thermal spreaderElectrical insulator, thermal spreader

Electrical insulator, thermal spreaderElectrical insulator, thermal spreader

Ther

mal

Inte

rfac

esTh

erm

al In

terf

aces SiGeSiGe

PbTePbTe

BiBi22

TeTe33

Heat ExchangerHeat Exchanger

Heat ExchangerHeat Exchanger

HOTHOT

COLDCOLD

(Morgan)(Morgan)

Research Challenge 4:

Electrical Contact Resistance

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 100 200 300 400 500 600

T

Effic

ienc

y,

Z=6E-3, Rc=0Z=6E-3, Rc=10E-8Z=6E-3, Rc=10E-7Z=6E-3, Rc=10E-6Z=6E-3, Rc=10E-5

•

Bad contacts can degrade perf.

•

Need: –

Cost-effective, producable–

temporally stable

•

Goal: Rc > 10-7

-cm2 (films)

•

Shoes/interconnect integration:–

Efficient heat transfer–

managing CTE mismatch–

manufacturable

For ZT ~ 2For ZT ~ 2

••

Contacts Contacts --

InterconnectsInterconnects

Take Aways…

•

ARL is acutely interested in TE technologies

•

Power Generation:–

Small power devices > 2 W/cm2

–

Thin-film Materials (MEMSUGS)Crystalline Bi2

Te3

/SiCrystalline (Pb,Sn)(Se,Te)/Si

–

Fast, simple

measusing radiative heat–

Seedling effort for CoSb3

•

Challenges/ Focus Areas:–

Higher Z–

Thermal management (heat sinking)–

Packaging–

Electrical contactsPartneringCollaboration

Back-up

975 W975 W--hr / kghr / kg

650 W650 W--hr / kghr / kg

325 W325 W--hr / kghr / kg

Goal:Goal:•

Develop small light weight power sources for the Warfighter

that maximize specific energy for Soldier systems and sensors.

•

High efficiency thermoelectrics

could compete with fuel cells, while likely using logistic fuels

Research Areas:Research Areas:•

Burner development•

TE materials•

TE packaging / interconnects•

Thermal management•

Balance of plant (pumps, valves, etc)

Direct Soldier Power Generation

JPJP--8 8 x x TE EfficiencyTE Efficiency

x System Overhead = x System Overhead = Energy DensityEnergy Density

15%15%

10%10%

5%5%

~50%~50%~13,000 ~13,000

WW--hr / kghr / kg