Solid State Chemistry Meets Physcis: Thermoelectric...

�

Solid State Chemistry Meets Physcis: Thermoelectric Materials

Solid State Chemistry Meets Physcis: Thermoelectric Materials

AN INTERDISCIPLINARY COLLABORATION

KANATZIDISMSU, CHEMISTRY

Solid State ChemistrySynthesis, Discovery

KANATZIDISMSU, CHEMISTRY

Solid State ChemistrySynthesis, Discovery

Ctirad UHERUniv of Michigan

Physics

Ctirad UHERUniv of Michigan

Physics

TIM HOGANMSU, E. ENGINEERING

MEASUREMENTS

S. D. MAHANTIMSU, PHYSICS

Theory

C. KANNEWURFNorthwestern Univ

E. Engineering

H. Schock, T. ShihMSU, Mechanical

EngineeringApplications

�

Michigan State University…Michigan State University…

�

THERMOELECTRIC POWER(Seebeck Coefficient)

ΔV Smeasured= ΔVΔT

Smeasured= Ssample-SCu

ΔT

Cu block

Cu blockheater

constantan

sample

zero current technique:extremely useful probefor investigation ofintrinsic conduction in granular or polycrystallinematerials

�

Thermoelectric ApplicationsThermoelectric Applications

Beverage cooler

Biological samples LASER diodeCooler

MicroprocessorCooler

www.tellurex.com

�

Heat to Electric EnergyHeat to Electric Energy

Up to 20% conversion efficiency withright materials

�

Thermoelectric BenefitsThermoelectric Benefits

• TE coolers have no moving parts, need substantially less maintenance.

• Life-testing has shown the capability of TE devices to exceed 200,000 hrs. of steady state operation.

• TE coolers contain no chlorofluorocarbons. • Temperature control to within fractions of a degree can

be maintained using TE devices. • TE coolers function in environments that are too

severe, too sensitive, or too small for conventional refrigeration.

• TE coolers are not position-dependent.

�

How does it work?How does it work?

http://www.designinsite.dk

�

Thermoelectric applicationsThermoelectric applications

• Air conditioning (distributed, environmentally friendly)

• Spot cooling of electronic chips, superconductors etc.

• Thermal suits for fire-fighting, soldier etc• Waste heat recovery (automobiles, utilities

etc)• Geothermal power generation

�

Figure of MeritFigure of Merit

ZT =σ ⋅ S2

κ total

•T

σ ⋅ S2Power factor

Total thermal conductivity

electrical conductivity thermopower

�

Today’s situationToday’s situation

• The most efficient materials today is Bi2Te3alloy

• ZT~0.8-1.0• Further improvements on Bi2Te3 are not

expected.• New materials are needed

�

Thermoelectric Properties of Optimized Bi2Te3(e.g. Bi2-xSbxTe3, Bi2Te3-xSex) at Room Temperature

Thermoelectric Properties of Optimized Bi2Te3(e.g. Bi2-xSbxTe3, Bi2Te3-xSex) at Room Temperature

• S ~ ±220 µV/K• σ ~ 950 S/cm• ρ=1/σ ~ 1.1 mΩ·cm• κ ~ 1.5 W/m·K

• ZT ~ 1 !

ZT~1

T, K

ZT

300 400200

�

Structure of Bi2Te3 and NaClStructure of Bi2Te3 and NaCl

xy

z

x

z

NaCl Bi2Te3 defect NaCl

�

TYPICAL BEHAVIOR OF MATERIALS

S>50 µV/K

S>-50 µV/K

0<S<20 µV/K

-20 µV/K<S<0

T (K)

S (µV/K)S (µV/K)

T (K)

metal-likesemiconductor-like

p-type

n-type n-type

p-type

LARGE

SMALL

�

Power Factor (S2*σ) vs Carrier Concentration

Power Factor (S2*σ) vs Carrier Concentration

Carrier Concentration, cm-1

1017 1018 1019 1020 1021

S2*σThermopower, Sconductivity

�

Thermopower and Electronic Structure

• σ(E) is the electrical conductivity determined as a function of band filling or Fermi energy, EF. If the electronic scattering is independent of energy, σ(E)is just proportional to the density of states (DOS) at EF.

• For maximum S, a large asymmetry in the DOS and/or scattering within a few kTabove and below the Fermi energy is required.

S =π2kB

2T3e

d(lnσ(E))dE

E = EFMott Equation

�

Band structure TypesBand structure Types

k

E

k

E

k

Ef

simple simple complex

�ZT and Band Structure

B =CT 5 / 2γ mxmym z

μx

κ latt

B- parameter

m= effective massµ= mobilityκlatt= lattice thermal conductivityT = temperatureγ= band degeneracy

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

ZT

B-parameter

1.0 -

10 –

0.1 -

1.0 100.10.01

Γ X

Ene

rgy

VB

CB

Ef

�

Desirable characteristicsDesirable characteristics

• Multiple peaks and valleys in valence/conduction band

• Heavy carrier masses• Flat bands

εε

k k

�

Selection criteria for candidate materials

Selection criteria for candidate materials

• Narrow band-gap semiconductors• For operation at room temperature

• Heavy elements• High mobility, low thermal conductivity

• Large unit cell, complex structure• low thermal conductivity

• Highly anisotropic or highly symmetric• Complex compositions

• low thermal conductivity, electronic structure

�

Important Issue: Thermal Conductivity

Important Issue: Thermal Conductivity

• Slack’s proposal: Phonon-Glass/ Electron-Crystal (PGEC)• Rattling Ions in the lattice: watch thermal

displacement parameters

o o o o

o o o oo

o

o

o

o

oooo o oooo o o

o o o o o

o

o

Rattling ions in cavities or tunnels scatter heat-carrying phonons

Crystalline solid

�

Reaction ChemistryReaction Chemistry

• A2Q + PbQ + Bi2Q3 –––––> (A2Q)n(PbQ)m(Bi2Q3)p

A2Q

PbQBi2Q3

A=K, Rb, CsQ=Se, Te

K5Bi17Se28

β-K2Bi8Se13

Rb0.5Bi1.83Te3

Pb5Bi6Se14

A1+xPb4-2xBi7+xSe15

APb2Bi3Te7

Pb6Bi2Se9 Pb5Bi12Se23

Map generates target compounds

CsBi4Te6

Investigating the System:

Phases shownare promising new TEMaterials

�

K ions possess ~2x the thermal displacement parameter of the Bi/Se framework

K2Bi8Se13

x y

z

Se

Bi

K

β-K2Bi8Se13this block is a chunk from the NaCl lattice

8,9-coordinate sites K, Bi

�

NaCl Structure: The Basic “Raw” Material

NaCl Structure: The Basic “Raw” Material

NaCl Lattice

“Modules” are cut out of NaCl stock

�

β-K2Bi8Se13β-K2Bi8Se13

0

20

40

60

80

100

0 50 100 150 200 250 300

Con

duct

ivity

(S/c

m)

Temperature (K)

-400

-300

-200

-100

0

0 50 100 150 200 250 300 350

Ther

mop

ower

(µV

/K)

Temperature (K)

-400

-350

-300

-250

-200

-150

-100

-50

0

0 50 100 150 200 250 300

Ther

mop

ower

(µV

/K)

Temperature (K)

150

200

250

300

350

400

0 50 100 150 200 250 300

Con

duct

ivity

(S/c

m)

Temperature (K)

Sn Doped

�

β-K2Bi8Se13 : Room temp ZT=0.9. At 600 K estimated at 1.5 (to be verified)

β-K2Bi8Se13 : Room temp ZT=0.9. At 600 K estimated at 1.5 (to be verified)

10

15

20

25

30

35

40

45

0.45 0.5 0.55 0.6 0.65 0.7 0.75

α/S

(arb

itrar

y un

it)

Eg = 0.59 eV

β-K2Bi

8Se

13

Energy (eV)

-250

-200

-150

-100

-50

0

0 50 100 150 200 250 300

β-K2Bi

8Se

13

Thermopower (μV/K)

Temperature (K)

DY61243 ingot

0

500

1000

1500

2000

2500

3000

0 50 100 150 200 250 300 350

Conductivity (S/cm)

Temperature (K)

0

1

2

3

4

5

6

0 50 100 150 200 250 300

β-K2Bi

8Se

13

κ (W

/m. K

)

Temperature (K)

DY61243 ingot

Uher et al

�

Sn Doping in β-K2Bi8Se13Sn Doping in β-K2Bi8Se13

-350

-300

-250

-200

-150

-100

-50

0

0 50 100 150 200 250 300

Undoped0.5% Sn1.0% Sn1.5% Sn2.0% Sn3.0% Sn

wer

(µV

/K)

Temperature (K)

0

5

10

15

20

25

30

35

40

0 50 100 150 200 250 300

Undoped0.5% Sn1.0% Sn1.5% Sn2.0% Sn3.0% Sn

ower

Fac

tor S

2 σ

(µW

/cm

·K

2)

Temperature (K)

0

0.2

0.4

0.6

0.8

1.0

0 50 100 150 200 250 300

Undoped0.5% Sn1.0% Sn1.5% Sn2.0% Sn3.0% Sn

ZT

Temperature (K)

�

Michigan State University /Tellurex Corp. Collaboration

Michigan State University /Tellurex Corp. Collaboration

β-K2Bi8Se13

New TE material grown at MSU

New TE material grown at Tellurex

�

Photo of the first TE module containing 63 couples n-β-K2Bi8Se13/p-Bi2Te3

Photo of the first TE module containing 63 couples n-β-K2Bi8Se13/p-Bi2Te3

UnoptimizedΔT=36 oCTh=50 oCAll materials grownat Tellurex Inc

�x

y

z xy

z

α-K2Bi8Se13 versus β-K2Bi8Se13α-K2Bi8Se13 versus β-K2Bi8Se13

α-K2Bi8Se13, Eg=0.76 eV β-K2Bi8Se13, Eg=0.59 eV

Mixed K,Bi

�

α-, β-K2Bi8Se13 : Electronic structureα-, β-K2Bi8Se13 : Electronic structure

n-type character

�

Quenched and annealed β-K2Bi8Se13Quenched and annealed β-K2Bi8Se13

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Abs

orpt

ion

Energy, eV

after annealingquenched

Eg=0.59 eV

�

CsBi4Te6CsBi4Te6

xy

z

C 2/m

51 Å

�

“Undoped” as-prepared material“Undoped” as-prepared material

0

20

40

60

80

100

120

0

5

10

15

20

0 50 100 150 200 250 300

S (µ

V/K

) κ (W/m

-K)

Temperature (K)

DY72121

0

50

100

150

200

250

300

350

0 50 100 150 200 250 300

dy72121(CsBi4Te6 crystal)

Con

d. (S

/cm

)

Temperature (K)

conductivity thermopower

Thermal conductivity

�

Crystals of CsBi4Te6Crystals of CsBi4Te6

1 mm1 mm

100 µm

�

Doped CsBi4Te6Doped CsBi4Te6

0

1000

2000

3000

4000

5000

6000

7000

8000

0

50

100

150

200

0 50 100 150 200 250 300 350

SbI3 doped CsBi

4Te

6

Con

duct

ivity

(S/c

m) S

eebeck (µV/K

)

Temperature (K)

At 260 K: S ~174 µV/KCond. ~1400 S/cm

κ ~14 mW/cm-K

ZTmax

~0.8

�

Thermal Conductivity of p-type CsBi4Te6

Thermal Conductivity of p-type CsBi4Te6

0.0

1.0

2.0

3.0

4.0

5.0

6.0

0 50 100 150 200 250 300 350

κ (W

/m. K

)

Temperature (K)

parallel to needle

perpendicular to needle

Data by Uher et al

�

Doping with SbI3Doping with SbI3

�

CsBi4-xSbxTe6 x = 0.3

SbSbBiBi

�

CsBi4Te6CsBi4Te6

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

0 50 100 150 200 250 300 350

Figure of Merit (ZT)

CsBi4Te

6 doped

p-Bi2Te

3 Marlow

ZT

Temperature (K)

�

Best TE MaterialsBest TE Materials

0

0.2

0.4

0.6

0.8

1

0 200 400 600 800 1000 1200 1400

n-BiSb �

n-SiGe sintered

CsBi4Te

6

LaFe3CoSb

12

Bi2Te

3

PbTe

ZT

Temperature (K)

RT

Temperature (K)

ZT

�

ConclusionsConclusions

• The strategy to search for new materials in the (A2Q)n(PbQ)m(Bi2Q3)p (Q=Se, Te) system is successful

• Many new promising compounds identified• All compounds strongly anisotropic• Doping studies are important in ZT optimization• ZT for β-K2Bi8Se13 ~0.7 at rt, higher at >400K

�

ReferencesReferences• “The Role of Solid State Chemistry In The Discovery of New

Thermoelectric Materials” Mercouri G. Kanatzidis, Semiconductors and Semimetals, 2000, 69, 51-100.

• Slack,G. A. “New Materials and Performance Limits for Thermoelectric Cooling” in CRC Handbook of Thermoelectrics" Edited by Rowe, D. M. CRC Press, Boca Raton, 1995, pp. 407-440

• Tritt T. M. "Thermoelectrics run hot and cold", Science. 1996, 272, 5266, 1276-1277.

• Mahan, G. D. “Good thermoelectrics” Solid State Phys: 1998, 51, 81-157. (c) DiSalvo, F. J. "Thermoelectric cooling and power generation", Science. 1999, 285 5428, 703-706

• Thermoelectric Materials 1998- The Next Generation Materials for Small-Scale Refridgeration and Power Generation Applications, edited by Tritt, T. M.; Kanatzidis, M. G.; Mahan, G. D.; Lyon, Jr., H. B. Mat. Res. Soc.Symp. Proc. 1999, Vol. 545, 233-246.

�

CollaboratorsCollaborators

Prof. Tim Hogan, Dept of Electrical Engineering, MSUProf. S. D. (Bhanu) Mahanti, Dept. of Physics, MSUProf. Carl R. Kannewurf, Dept of Electrical Engineering, Northwestern Univ.Ctirad Uher, Dept. of Physics, U of MArt Schultz, Argonne NL

�

TE Research groupTE Research group

• Dr Duck young Chung• Dr Antje Mrotzek• Dr Kuei fang Hsu• Lykourgos Iordanidis• Kyoung-shin Choi• Jun-Ho Kim• Sandrine Sportouch• Rhonda Patschke

• Tim McCarthy• Dr. Jun-Huan Do

�

AcknowledgementsAcknowledgements• Dr. Antje Mrotzek• Lykourgos Iordanidis• J. A. Aitken• Jun-Ho Kim• Joseph Wachter• Marina Zhuravleva• Xuini Wu• Jim Salvador• Brad Sieve• R G Iyer• Dr. Theodora Kyratsi• Dr. J.-H. Do• Dr. Duck Young Chung• Dr. Servane Coste• Dr. Pantelis Trikalitis• Dr. Susan Latturner• Dr. Kuei-fang Hsu