From Heat to Electricity: How “nano”Saved Thermoelectrics

Sponsored by

Mercouri Kanatzidis, Northwestern University

Collaborating Team interactions

Kanatzidis groupChemistry

Materials design, discoveryand optimization

Hogan groupE. Engineering

Module design and optimization

Uher groupPhysics

TE Measurements

Mahanti groupPhysics (theory)

Fundamental understandingElectronic structure, Property prediction

Schock groupM. Engineering

Materials scale-up andoptimization, design/testing

Case groupMaterials Science

Powder processingMechanical properties

Tellurex CorpIndustry

Bruce CookTEM, transport

Outline

IntroductionMaterials under investigationSynthesis (successes and challenges)TEM characterization and nanostructuringStructure property relationships- Nanostructures reduce the lattice thermal conductivityMaterials scale up and optimizationModule Fabrication Progress, low resistance contactsMaking brittle materials strongConclusions

Heat to Electrical Energy DirectlyUp to 20% conversion efficiency with right materials

http://www.dts-generator.com/

TE devices have no moving parts, no noise, reliable

Thermoelectric applicationsWaste heat recovery • Automobiles• Over the road trucks• Utilities• Chemical plants

Space powerRemote Power GenerationSolar energyGeothermal power generationDirect nuclear to electrical

U.S. Energy Flow, 1999

web site: www.eia.doe.gov

Given that ~60% of energy becomes waste heat, even a 10% capture and conversionto useful forms can have huge impact on overall energy utilization

Figure of Merit

ZT =σ ⋅ S2

κ total

•T

σ ⋅ S2Power factor

Total thermal conductivity

electrical conductivity thermopower

ZT =σ ⋅ S2

κ total

•T

σ ⋅ S2Power factor

Total thermal conductivity

electrical conductivity thermopower

0

0.05

0.1

0.15

0.2

0.25

0 0.5 1 1.5 2 2.5 3

η

ZTavg

δ = 0.0δ = 0.1

δ = 1.1

0

0.05

0.1

0.15

0.2

0.25

0 0.5 1 1.5 2 2.5 3

η

ZTavg

δ = 0.0δ = 0.1

δ = 1.1

δ=Rc/RFor Th = 800K

Tc = 300K

Today’s situationThe most efficient materials today for power generation:PbTe: ZT~0.8 at 800 K (n-type)TAGS: ZT~1.2 700 K (p-type)Bi2Te3-xSex: ZT~1 at 300 KFurther improvements are needed.New materials emerging

PbTePbSe 20 nm dot

Quantum Dot Layers in thin MBE-grown PbSe/PbTe superlattices (Harman et al, ZT~3)

10Å/50Å Bi2Te3/Sb2Te3 superlattice structures

Venkatasubramanian R, Siivola E, Colpitts T et al. Nature, 2001, 413: 597

Harman T C, Taylor P J, Walsh M P et al. Science, 2002, 297: 2229

Some promising systems under investigation

half-Heusler alloys (ZrNiSn)Zn4Sb3ClathratesSkutterudites (CoSb3)Bulk nanocomposites based on PbTe Bulk nanocomposites based on Si-GeAgSbTe2/PbTe, NaSbTe2/PbTe

March 2006 issue of MRS Bulletin

( )2/1

2/3

max+∝ r

latt

z

yx

em

mmT

Zκ

τγ

ZT and Electronic Structure

m= effective massτ=scattering timer= scattering parameterκlatt= lattice thermal conductivityT = temperatureγ= band degeneracy

Large γ comes with(a) high symmetry e.g. rhombohedral, cubic(b) off-center band extrema

Isotropic structure

Anisotropic structure

k

E

k

E

k

Ef

For acoustic phonon scatteringr=-1/2

Complex electronic structure

Selection criteria for candidate materials

Narrow band-gap semiconductorsHeavy elements

High μ, low κLarge unit cell, complex structure

low κHighly anisotropic or highly symmetric…Complex compositions

low κ, complex electronic structure

A2Q + PbQ + M2Q3 –––––> (A2Q)n(PbQ)m(M2Q3)p

Investigating the A/Bi/Q system

A2Q

PbQBi2Q3

A=Ag, K, Rb, CsM=Sb, BiQ=Se, Te

β-K2Bi8Se13

Rb0.5Bi1.83Te3

Pb5Bi6Se14

A1+xPb4-2xBi7+xSe15

APb2Bi3Te7

Pb6Bi2Se9 Pb5Bi12Se23

Map generates target compounds

CsBi4Te6

Phases shownare promising new TEmaterials

Cubic materialsAmBnMmQ2m+n

Compounds discovered

K2Bi8Se13, KPbBi9Se13, KPb4Sb7Se15

Cs1-xPb5-xBi10+xSe21

CsPbBi3Te6, CsPb2Bi3Te7, RbPbBi3Te6, RbPb2Bi3Te7, RbPb3Bi3Te8,KPbBiSe3, K2PbBi2Se5

AgPb10SbTe12, NaPb20SbTe22

AgPbmSbTe2+m (LAST-m) NaPbmSbTe2+m (SALT-m)

AgSbTe

Pb

6 8 10 12 14 16 18

263

264

265

266

267

268

Uni

t cel

l vol

ume

(Å)

Vegard's law

m value

Powder data

Single crystal data

(1) (a) Rodot, H. Compt. Rend. 1959, 249, 1872-4. (2) (a) Rosi, F. D.; Hockings, E. S.; Lindenblad, N. E. Adv. EnergyConvers. 1961, 1, 151.

No phase transitions to melting point

Synthesis

R. G. Maier Z. Metallkunde 1963, 311

T, K

time, h

1000 ºC

Cool to 50 ºCin 24 h

Gravity induced inhomogeneityLAST-18

time, h

T, K950-1000 ºC Cool to 50 ºC in

72 h

13 deg/h

Ingot properties very sensitive to cooling profile

Wernick, J. H.. Metallurg. Soc. Conf. Proc. (1960), 5 69-87.

rock

rock

Large scale LAST-18 (n-type)

Kosuga et al J. Alloys Compnds, 2005, 391, 288

0

0.5

1

1.5

2

320 400 480 560 640 720 800 880

ZT

Temperature (K)

LAST-18

ETN-16 (LAST-18)

LAST-18Kosuga et al 2005

PbTe

LAST-22Wang et al 2006

m= 40

m= 25

m= 18

m= 14m= 8

Strong composition grading along ingot

Deviations from the AgPb18SbTe20 compositionBest samples are Te-deficient

Ag0.86Pb19SbTe20EDS analysis

49.8247.201.391.59Average

49.7147.431.441.42

49.6547.181.471.70

49.7547.271.391.60

50.1746.921.261.65

TePbSbAgETN125A

49.4447.561.431.57Average

49.5146.981.472.04

49.7047.361.761.18

49.0847.651.451.82

48.9448.181.261.61

49.9847.631.211.19

TePbSbAgETN125B

49.7847.451.561.21Average

49.9647.351.790.89

49.5048.051.720.73

49.8047.351.111.74

49.8747.041.601.49

TePbSbAgETN125C

49.5347.891.131.50Average

49.9747.731.181.12

48.9247.061.362.65

49.6147.701.241.45

49.5648.021.391.02

49.9248.190.851.55

TePbSbAgETN125D

7 d4 h1000 oC

700 oC

12 h

50 oC

12 h

LAST-18: Synthesis with Slow Cooling

fast cooled sample

slow cooled sample

105 g201910.86amountTePbSbAg

25.6-1581026C

-180

-128

-121

S(μV/K)

43.41341D

15.7959B

7.8535A

PF(μW/cm∙K2)

σ(S/cm)

ETN125

~2deg/hr

Properties of Ag1-xPb18SbTe20

0

500

1000

1500

2000

-400

-350

-300

-250

-200

-150

-100

300 350 400 450 500 550 600 650 700

σ (S

/cm

) S (μV

/K)

Temperature, K0.00

0.50

1.00

1.50

2.00

300 350 400 450 500 550 600 650 700

Ag0.86

Pb18

SbTe20

ZT

Temperature, K

PbTe

TEM investigation on nanostructuring: LAST-18 (Cook)

Nanostructures reduce the lattice thermal conductivity

0.35 W/mK (Harman PbTe/PbSe superlattice)0

0.5

1

1.5

2

2.5

300 400 500 600 700 800

Latti

ce T

herm

al C

ondu

ctiv

ity (W

/mK

)

Temperature (K)

PbTe

LAST-18

Lattice thermal conductivity

Clemens-Drabble theory

Recent Report for Sintered Pellet Ag0.8Pb22SbTe20, ZT ~ 1.4 at 680 K

P-type materials, LASTT(LASTT-m) Ag(Pb1-xSnx)mSbTe2+mSn atoms act as acceptorsAg atoms act as acceptorsSb atoms act as donorse.g AgPb10Sn8SbTe20, AgxPb7Sn3SbyTe12, Ag1-xSnSb1+xTe3, etcVery low lattice thermal conductivityGood homogeneity

Hogan

PF~17-18 μW/cmK2

200

400

600

800

1000

300 400 500 600 700

Ele

ctri

cal C

ondu

ctiv

ity (S

/cm

)

Temperature (K)

Ag0.5

Pb6Sn

2Sb

0.2Te

10

50

100

150

200

250

300

300 400 500 600 700

The

rmop

ower

(µV

/K)

Temperature (K)

Ag0.5

Pb6Sn

2Sb

0.2Te

10

LASTT: Very low lattice thermal conductivity

J. Androulakis, K. F. Hsu, R. Pcionek, H. Kong, C. Uher, J. J. D'Angelo, A. Downey, T. Hogan, M. G. Kanatzidis, Advanced Materials 2006, 18, 1170

300 350 400 450 500 550 600 650 700 7500.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

κ (W

/mK

)

Temperature (K)

κlatt, PbTe

κlatt,LASTT

LASTT-16: AgPb14Sn4Sb0.4Te20

LASTT-10: AgPb10Sn10Sb0.7Te22

0 100 200 300 400 500 6000.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

0.018

JAE69P110-BAg0.5Pb6Sn2Sb0.2Te10

κ (W

/cm

K)

Temperature (K)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

κ (W

/mK)

0 100 200 300 400 500 6000.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

0.018

JAE69P110-BAg0.5Pb6Sn2Sb0.2Te10

κ (W

/cm

K)

Temperature (K)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

κ (W

/mK)

Why do the LAST materials nanostructure?

Pb Te Pb Te Pb Te Pb

Te Pb Te Pb Te Pb Te

Pb Te Sb Te Pb Te Pb

Te Ag Te Ag Te Pb Te

Pb

Te Pb

Te

Te

Sb

Pb

Te

Te

Pb

Pb

Te

Te

Pb

Ag Te Pb Te Pb Te Pb

Te Pb Te Pb Te Sb Te

Pb Te Pb Te Pb Te Pb

Te Ag Te Pb Te Pb Te

Pb

Te Pb

Te

Te

Pb

Pb

Te

Te

Sb

Pb

Te

Te

Pb

Dissociated state..unstable Associated state..stable

Any +1/+3 pair

Driving force for segregation Ag+/Sb3+ pair: thermodynamics

Figure of Merit LASTT (p-type)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8 Ag0.9Pb10.8Sn7.2Sb0.6Te20 Ag0.9Pb7.2Sn10.8Sb0.6Te20 Ag0.9Pb9Sn9Sb0.6Te20 Ag0.6Pb7Sn3Sb0.2Te12 Ag0.5Pb6Sn2Sb0.2Te10Ag0.9Pb5Sn3Sb0.7Te10

ZT

(a)

J. Androulakis, K. F. Hsu, R. Pcionek, H. Kong, C. Uher, J. J. D'Angelo, A. Downey, T. Hogan, M. G. Kanatzidis, Advanced Materials 2006, 18, 1170

Na-based materials (SALT-m)New high ZT p-type material Na1-xPbmSbTe2+m

m~19-21

5 nm

2 nm

What is nanostructuring worth?

0

0.5

1

1.5

2

300 350 400 450 500 550 600 650 700

Latti

ce T

herm

al c

ondu

ctiv

ity (W

/mK

)

Temperature, K

PbTe0.9

Se0.1

Pb0.9

Sn0.1

Te

Na1-x

Pb20

SbTe22

(B)

P. F. P. Poudeu, J. D'Angelo, A. D. Downey, J. L. Short, T. P. Hogan, M. G. Kanatzidis, Angew. Chem. Int. Ed. 2006, 45, 1

Pb Te Pb Te Pb Te Pb

Te Pb Te Pb Te Pb Te

Pb Te Sb Te Pb Te Pb

Te Na Te Na Te Pb Te

Pb

Te Pb

Te

Te

Sb

Pb

Te

Te

Pb

Pb

Te

Te

Pb

Matrix Encapsulation as a Route to Nanostructured PbTe

PbTe + X

X = InSb

X = Sb

X = B

i 20 nm

100 nm

2 nm

2 nm

100 nm

2 nm

Nanocrystals of Sb in PbTe

300 400 500 600 7000.5

1.0

1.5

2.0

2.5

PbTe - Sb(16%)PbTe - Sb(4%)

κ lat, W

m-1K

-1

Temperature, K

PbTe - Sb(2%)

PbTe

300 400 500 600 7000.5

1.0

1.5

2.0

2.5PbTe

4% InSb

4% Bi

κ lat,

Wm

-1K

-1

Temperature, K

4% Sb

• An optimum concentration of nanoscale second phase is necessary• Mass fluctuations play a role in thermal conductivity reduction• Lattice thermal conductivity reduced, however ZT low due to small Seebeck

20 nm

Phonons

Electrons

Completed and Processed IngotComposition: Ag0.43Pb18Sb1.2Te20 Weight: 200 grams

0

200

400

600

800

1000

1200

1400

1600

-280

-260

-240

-220

-200

-180

-160

-140

-120

350 400 450 500 550 600 650

1 - Increasing T2 - Decreasing T3 - Increasing T4 - Decreasing T

1 - Increasing T2 - Decreasing T3 - Increasing T4 - Decreasing T

Ele

ctric

al C

ondu

ctiv

ity (S

/cm

)

TEP (µV/K)

Temperature (K)

Ag0.43

Pb18

Sb1.2

Te20

Temperature cyclability

Schock

ETN611020oC to 700oC @ 120oC/hr

ETN641020oC to 825oC @ 97.5oC/hr

Cool from 1020ºC to 700, 800, 825, 850, 875, 900°C typically from one – two hours

Heating and Cooling Optimization of LASTAg0.43Pb18Sb1.2Te20

ETN651020oC to 875oC @ 72.5oC/hr

ETN681020oC to 900oC @ 120oC/hr

Module Fabrication

1.78mΩ total 16.0µΩ·cm2

Hot side diffusion contacts, and cold side solder contacts with <10 µW·cm2 have been achieved.

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

Th=495KTh=545KTh=585K

Pow

er G

ener

ated

(W)

Load resistance (ohms)

0

0.2

0.4

0.6

0.8

1

0 2 4 6 8 10

Vol

tage

(mV

)

Position (mm)

ETN123

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 1 2 3 4 5 6 7 8

Vol

tage

(mV

)

Position (mm)

Ni plated samples.SnPb to Ni.Kester Flux.Hot plate.

Scanning Probe Results

ETN123

crack

0

50

100

150

200

0 2 4 6 8 10

Vol

tage

(µV

)

Position (mm)

Ag0.9

Pb9Sn

9Sb

0.6Te

20

Jarrod Short

Ni

Ni

I = 100mASlope = 26 (µV/mm)4.7mm x 4.7mmσ = 1740 (S/cm)

Ni electrode on LASTT

In region 1 (large grains), grain size dominates the

fracture behavior

In region 2 (smaller grains), grain size and surface

defects contribute to fracture

For brittle materials in general, the fracture strength is a function of the reciprocal square root of grain size.

(Grain Size)-1/2

Stre

ngth

Grain Size (G)

Region 1

Region 2

Eldon CaseMaking brittle materials strong

Hot pressing

Powder Processing and Mechanical Characterization of LAST/T Materials

Ingot LAST/T materials have a large mean grain size (~ 500 to 700 μm), resulting in poor mechanical properties. We utilize planetary ball milling to generate micron-scale powders from LAST/T ingots. To date, we have achieved equiaxed powders with a mean particle size of ~ 5 μm via dry milling. Wet milling studies demonstrate further size reduction with a mean grain sizes of ~ 2 to 3 μm.

Eldon Case

To improve mechanical integrity, smaller grain size is needed.

Hot Press BilletsHPMSU-01 HPMSU-02

Schock, Case

1 inch

Best ZT Materials

0

0.5

1

1.5

2

0 200 400 600 800 1000 1200 1400

Temperature (K)

CsBi4Te

6

Bi2Te

3

Tl9BiTe

6

LAST

LASTTNa

0.95Pb

20SbTe

22

TAGS

Ce0.9

Fe3CoSb

12PbTe SiGeβ -K2Bi

8Se

13

ConclusionsLAST, LASTT and SALT: promising thermoelectric materials for next generation power generation modules. (expected device efficiency ~14%)Nanostructures strongly reduce thermal conductivity.Nanostructures are closely linked to high ZT.Scaleup successful in producing large quantities but material is brittle and contains microcracks.Hot pressing and powder processing yield 3x improvement in strength.Higher average ZT (>2) needed to reach 20% efficiency.

Students and postdocsGraduate Students

Joe Sootsman, Chemistry Huijun Kong, Physics, U. of MichiganAdam Downey, Electrical EngineeringJarrod Short, Electrical EngineeringJonathan D’Angelo, Electrical EngineeringFei Ren, Chem. Engineering and Materials ScienceChris Malliakas, ChemistryKhang Hoang, PhysicsAhmed Salameh, PhysicsMayank Mittal, Mech. EngineeringAurelie Guegen, Chemistry

UndergraduatesAdam Pilchak, Materials ScienceTeresa Rhodes, ChemistryJason Johnson, Materials Science

Postdocs and Research AssociatesFerdinand Poudeu, ChemistryEd Timm, Mechanical EngineeringRobert Pcionek, ChemistryChun-I Wu, E. EngineeringJim Salvador, ChemistryChia-Her Lin, ChemistryXun Shi, Physics U of Michigan

Sponsored by